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": "2",
"budget": 2142
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000460",
"normalized_example_id": "ex_000460",
"source_idx": 460,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2677,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2",
"budget": 2677
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000460",
"normalized_example_id": "ex_000460",
"source_idx": 460,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2677,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;2;1)",
"budget": 2828
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000461",
"normalized_example_id": "ex_000461",
"source_idx": 461,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4714,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;2;1)",
"budget": 3771
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000461",
"normalized_example_id": "ex_000461",
"source_idx": 461,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4714,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;2;1)",
"budget": 4714
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000461",
"normalized_example_id": "ex_000461",
"source_idx": 461,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4714,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{14}}{4}",
"budget": 2590
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000462",
"normalized_example_id": "ex_000462",
"source_idx": 462,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4316,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{14}}{4}",
"budget": 3453
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000462",
"normalized_example_id": "ex_000462",
"source_idx": 462,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4316,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{14}}{4}",
"budget": 4316
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000462",
"normalized_example_id": "ex_000462",
"source_idx": 462,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4316,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "22",
"budget": 4064
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000463",
"normalized_example_id": "ex_000463",
"source_idx": 463,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6773,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "22",
"budget": 5418
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000463",
"normalized_example_id": "ex_000463",
"source_idx": 463,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6773,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "22",
"budget": 6773
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000463",
"normalized_example_id": "ex_000463",
"source_idx": 463,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6773,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\{(-1, 0), (0, -1), (0, 1)\\}",
"budget": 4247
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000464",
"normalized_example_id": "ex_000464",
"source_idx": 464,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7078,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\{(-1, 0), (0, -1), (0, 1)\\}",
"budget": 5662
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000464",
"normalized_example_id": "ex_000464",
"source_idx": 464,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7078,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\{(-1, 0), (0, -1), (0, 1)\\}",
"budget": 7078
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000464",
"normalized_example_id": "ex_000464",
"source_idx": 464,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7078,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2369
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000466",
"normalized_example_id": "ex_000466",
"source_idx": 466,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3948,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3158
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000466",
"normalized_example_id": "ex_000466",
"source_idx": 466,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3948,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3948
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000466",
"normalized_example_id": "ex_000466",
"source_idx": 466,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3948,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(x, y, z) \\in \\{(1, 0, 2), (1, 2, 0)\\}",
"budget": 2409
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000467",
"normalized_example_id": "ex_000467",
"source_idx": 467,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4015,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(x, y, z) \\in \\{(1, 0, 2), (1, 2, 0)\\}",
"budget": 3212
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000467",
"normalized_example_id": "ex_000467",
"source_idx": 467,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4015,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(x, y, z) \\in \\{(1, 0, 2), (1, 2, 0)\\}",
"budget": 4015
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000467",
"normalized_example_id": "ex_000467",
"source_idx": 467,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4015,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\sqrt{2}-1",
"budget": 2752
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000468",
"normalized_example_id": "ex_000468",
"source_idx": 468,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4586,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\sqrt{2}-1",
"budget": 3669
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000468",
"normalized_example_id": "ex_000468",
"source_idx": 468,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4586,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\sqrt{2}-1",
"budget": 4586
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000468",
"normalized_example_id": "ex_000468",
"source_idx": 468,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4586,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(0,0,0),(0,1,-1),(-1,0,1),(1,-1,0)",
"budget": 5852
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000469",
"normalized_example_id": "ex_000469",
"source_idx": 469,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9754,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(0,0,0),(0,1,-1),(-1,0,1),(1,-1,0)",
"budget": 7803
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000469",
"normalized_example_id": "ex_000469",
"source_idx": 469,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9754,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(0,0,0),(0,1,-1),(-1,0,1),(1,-1,0)",
"budget": 9754
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000469",
"normalized_example_id": "ex_000469",
"source_idx": 469,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9754,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "218,219,220,221,222,224,225,226,227",
"budget": 1034
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000471",
"normalized_example_id": "ex_000471",
"source_idx": 471,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1723,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "218,219,220,221,222,224,225,226,227",
"budget": 1378
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000471",
"normalized_example_id": "ex_000471",
"source_idx": 471,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1723,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "218,219,220,221,222,224,225,226,227",
"budget": 1723
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000471",
"normalized_example_id": "ex_000471",
"source_idx": 471,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1723,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1202
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000473",
"normalized_example_id": "ex_000473",
"source_idx": 473,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2003,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1602
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000473",
"normalized_example_id": "ex_000473",
"source_idx": 473,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2003,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 2003
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000473",
"normalized_example_id": "ex_000473",
"source_idx": 473,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2003,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{16}",
"budget": 1054
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000476",
"normalized_example_id": "ex_000476",
"source_idx": 476,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1757,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{16}",
"budget": 1406
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000476",
"normalized_example_id": "ex_000476",
"source_idx": 476,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1757,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{16}",
"budget": 1757
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000476",
"normalized_example_id": "ex_000476",
"source_idx": 476,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1757,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "P(x) = 0 \\text{ or } P(x) = x^2 + x",
"budget": 3067
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000477",
"normalized_example_id": "ex_000477",
"source_idx": 477,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5111,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "P(x) = 0 \\text{ or } P(x) = x^2 + x",
"budget": 4089
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000477",
"normalized_example_id": "ex_000477",
"source_idx": 477,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5111,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "P(x) = 0 \\text{ or } P(x) = x^2 + x",
"budget": 5111
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000477",
"normalized_example_id": "ex_000477",
"source_idx": 477,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5111,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\pi}{4}",
"budget": 3672
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000478",
"normalized_example_id": "ex_000478",
"source_idx": 478,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6120,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{\\pi}{4}",
"budget": 4896
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000478",
"normalized_example_id": "ex_000478",
"source_idx": 478,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6120,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{\\pi}{4}",
"budget": 6120
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000478",
"normalized_example_id": "ex_000478",
"source_idx": 478,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6120,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "358926471",
"budget": 5149
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000479",
"normalized_example_id": "ex_000479",
"source_idx": 479,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8581,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "358926471",
"budget": 6865
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000479",
"normalized_example_id": "ex_000479",
"source_idx": 479,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8581,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "358926471",
"budget": 8581
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000479",
"normalized_example_id": "ex_000479",
"source_idx": 479,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8581,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "45",
"budget": 5514
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000481",
"normalized_example_id": "ex_000481",
"source_idx": 481,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9190,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "45",
"budget": 7352
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000481",
"normalized_example_id": "ex_000481",
"source_idx": 481,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9190,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "45",
"budget": 9190
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000481",
"normalized_example_id": "ex_000481",
"source_idx": 481,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9190,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1849
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000482",
"normalized_example_id": "ex_000482",
"source_idx": 482,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3081,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2465
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000482",
"normalized_example_id": "ex_000482",
"source_idx": 482,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3081,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3081
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000482",
"normalized_example_id": "ex_000482",
"source_idx": 482,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3081,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4347",
"budget": 1354
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000484",
"normalized_example_id": "ex_000484",
"source_idx": 484,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2256,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4347",
"budget": 1805
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000484",
"normalized_example_id": "ex_000484",
"source_idx": 484,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2256,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4347",
"budget": 2256
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000484",
"normalized_example_id": "ex_000484",
"source_idx": 484,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2256,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2005
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000486",
"normalized_example_id": "ex_000486",
"source_idx": 486,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3341,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2673
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000486",
"normalized_example_id": "ex_000486",
"source_idx": 486,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3341,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3341
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000486",
"normalized_example_id": "ex_000486",
"source_idx": 486,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3341,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "167",
"budget": 4820
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000489",
"normalized_example_id": "ex_000489",
"source_idx": 489,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8033,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "167",
"budget": 6426
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000489",
"normalized_example_id": "ex_000489",
"source_idx": 489,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8033,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "167",
"budget": 8033
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000489",
"normalized_example_id": "ex_000489",
"source_idx": 489,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8033,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "491",
"budget": 2824
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000490",
"normalized_example_id": "ex_000490",
"source_idx": 490,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4707,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "491",
"budget": 3766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000490",
"normalized_example_id": "ex_000490",
"source_idx": 490,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4707,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "491",
"budget": 4707
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000490",
"normalized_example_id": "ex_000490",
"source_idx": 490,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4707,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{5}{2}",
"budget": 3735
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000492",
"normalized_example_id": "ex_000492",
"source_idx": 492,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6225,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{5}{2}",
"budget": 4980
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000492",
"normalized_example_id": "ex_000492",
"source_idx": 492,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6225,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{5}{2}",
"budget": 6225
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000492",
"normalized_example_id": "ex_000492",
"source_idx": 492,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6225,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\ln3",
"budget": 2630
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000493",
"normalized_example_id": "ex_000493",
"source_idx": 493,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4383,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\ln3",
"budget": 3506
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000493",
"normalized_example_id": "ex_000493",
"source_idx": 493,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4383,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\ln3",
"budget": 4383
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000493",
"normalized_example_id": "ex_000493",
"source_idx": 493,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4383,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{a l}+\\sqrt{b m}+\\sqrt{c n})^{2}}{\\delta}",
"budget": 3110
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000494",
"normalized_example_id": "ex_000494",
"source_idx": 494,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5184,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{a l}+\\sqrt{b m}+\\sqrt{c n})^{2}}{\\delta}",
"budget": 4147
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000494",
"normalized_example_id": "ex_000494",
"source_idx": 494,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5184,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{a l}+\\sqrt{b m}+\\sqrt{c n})^{2}}{\\delta}",
"budget": 5184
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000494",
"normalized_example_id": "ex_000494",
"source_idx": 494,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5184,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{n}{n-1}",
"budget": 2907
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000496",
"normalized_example_id": "ex_000496",
"source_idx": 496,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4845,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{n}{n-1}",
"budget": 3876
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000496",
"normalized_example_id": "ex_000496",
"source_idx": 496,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4845,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{n}{n-1}",
"budget": 4845
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000496",
"normalized_example_id": "ex_000496",
"source_idx": 496,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4845,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4,\\frac{3\\sqrt{5}}{2}",
"budget": 2420
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000499",
"normalized_example_id": "ex_000499",
"source_idx": 499,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4034,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4,\\frac{3\\sqrt{5}}{2}",
"budget": 3227
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000499",
"normalized_example_id": "ex_000499",
"source_idx": 499,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4034,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4,\\frac{3\\sqrt{5}}{2}",
"budget": 4034
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000499",
"normalized_example_id": "ex_000499",
"source_idx": 499,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4034,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "250",
"budget": 1829
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000500",
"normalized_example_id": "ex_000500",
"source_idx": 500,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3049,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "250",
"budget": 2439
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000500",
"normalized_example_id": "ex_000500",
"source_idx": 500,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3049,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "250",
"budget": 3049
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000500",
"normalized_example_id": "ex_000500",
"source_idx": 500,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3049,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\pi}{4}",
"budget": 2091
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000501",
"normalized_example_id": "ex_000501",
"source_idx": 501,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3485,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\pi}{4}",
"budget": 2788
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000501",
"normalized_example_id": "ex_000501",
"source_idx": 501,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3485,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\pi}{4}",
"budget": 3485
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000501",
"normalized_example_id": "ex_000501",
"source_idx": 501,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3485,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 5096
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000503",
"normalized_example_id": "ex_000503",
"source_idx": 503,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8493,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 6794
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000503",
"normalized_example_id": "ex_000503",
"source_idx": 503,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8493,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 8493
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000503",
"normalized_example_id": "ex_000503",
"source_idx": 503,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8493,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4^{x}\\ln4",
"budget": 953
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000505",
"normalized_example_id": "ex_000505",
"source_idx": 505,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1588,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4^{x}\\ln4",
"budget": 1270
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000505",
"normalized_example_id": "ex_000505",
"source_idx": 505,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1588,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4^{x}\\ln4",
"budget": 1588
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000505",
"normalized_example_id": "ex_000505",
"source_idx": 505,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1588,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1496
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000506",
"normalized_example_id": "ex_000506",
"source_idx": 506,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2493,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1994
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000506",
"normalized_example_id": "ex_000506",
"source_idx": 506,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2493,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2493
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000506",
"normalized_example_id": "ex_000506",
"source_idx": 506,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2493,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{7}{27}",
"budget": 3955
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000507",
"normalized_example_id": "ex_000507",
"source_idx": 507,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6592,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{7}{27}",
"budget": 5274
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000507",
"normalized_example_id": "ex_000507",
"source_idx": 507,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6592,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{7}{27}",
"budget": 6592
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000507",
"normalized_example_id": "ex_000507",
"source_idx": 507,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6592,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 825
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000511",
"normalized_example_id": "ex_000511",
"source_idx": 511,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1375,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1100
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000511",
"normalized_example_id": "ex_000511",
"source_idx": 511,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1375,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1375
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000511",
"normalized_example_id": "ex_000511",
"source_idx": 511,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1375,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(3,3,4)",
"budget": 5989
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000514",
"normalized_example_id": "ex_000514",
"source_idx": 514,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9982,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(3,3,4)",
"budget": 7986
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000514",
"normalized_example_id": "ex_000514",
"source_idx": 514,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9982,
"... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.