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": "13",
"budget": 4788
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004591",
"normalized_example_id": "ex_004591",
"source_idx": 4591,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4788,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n\\geqslant2",
"budget": 1549
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004593",
"normalized_example_id": "ex_004593",
"source_idx": 4593,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2582,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n\\geqslant2",
"budget": 2066
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004593",
"normalized_example_id": "ex_004593",
"source_idx": 4593,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2582,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n\\geqslant2",
"budget": 2582
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004593",
"normalized_example_id": "ex_004593",
"source_idx": 4593,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2582,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\ln3}{2}",
"budget": 1043
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004594",
"normalized_example_id": "ex_004594",
"source_idx": 4594,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1738,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\ln3}{2}",
"budget": 1390
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004594",
"normalized_example_id": "ex_004594",
"source_idx": 4594,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1738,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\ln3}{2}",
"budget": 1738
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004594",
"normalized_example_id": "ex_004594",
"source_idx": 4594,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1738,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{2\\pi}",
"budget": 2413
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004596",
"normalized_example_id": "ex_004596",
"source_idx": 4596,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4022,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{2\\pi}",
"budget": 3218
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004596",
"normalized_example_id": "ex_004596",
"source_idx": 4596,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4022,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{2\\pi}",
"budget": 4022
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004596",
"normalized_example_id": "ex_004596",
"source_idx": 4596,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4022,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "r,\\frac{4}{3}r,\\frac{5}{3}r",
"budget": 4164
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004598",
"normalized_example_id": "ex_004598",
"source_idx": 4598,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6940,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "r,\\frac{4}{3}r,\\frac{5}{3}r",
"budget": 5552
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004598",
"normalized_example_id": "ex_004598",
"source_idx": 4598,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6940,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "r,\\frac{4}{3}r,\\frac{5}{3}r",
"budget": 6940
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004598",
"normalized_example_id": "ex_004598",
"source_idx": 4598,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6940,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 1132
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004601",
"normalized_example_id": "ex_004601",
"source_idx": 4601,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1887,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 1510
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004601",
"normalized_example_id": "ex_004601",
"source_idx": 4601,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1887,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 1887
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004601",
"normalized_example_id": "ex_004601",
"source_idx": 4601,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1887,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{2\\sqrt{6}}",
"budget": 1867
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004603",
"normalized_example_id": "ex_004603",
"source_idx": 4603,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3112,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{2\\sqrt{6}}",
"budget": 2490
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004603",
"normalized_example_id": "ex_004603",
"source_idx": 4603,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3112,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{2\\sqrt{6}}",
"budget": 3112
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004603",
"normalized_example_id": "ex_004603",
"source_idx": 4603,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3112,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "67",
"budget": 3348
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004605",
"normalized_example_id": "ex_004605",
"source_idx": 4605,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5580,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "67",
"budget": 4464
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004605",
"normalized_example_id": "ex_004605",
"source_idx": 4605,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5580,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "67",
"budget": 5580
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004605",
"normalized_example_id": "ex_004605",
"source_idx": 4605,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5580,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14",
"budget": 4129
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004607",
"normalized_example_id": "ex_004607",
"source_idx": 4607,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6881,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14",
"budget": 5505
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004607",
"normalized_example_id": "ex_004607",
"source_idx": 4607,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6881,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14",
"budget": 6881
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004607",
"normalized_example_id": "ex_004607",
"source_idx": 4607,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6881,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "59",
"budget": 1918
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004608",
"normalized_example_id": "ex_004608",
"source_idx": 4608,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3196,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "59",
"budget": 2557
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004608",
"normalized_example_id": "ex_004608",
"source_idx": 4608,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3196,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "59",
"budget": 3196
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004608",
"normalized_example_id": "ex_004608",
"source_idx": 4608,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3196,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{2008}-2^{2008}",
"budget": 945
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004612",
"normalized_example_id": "ex_004612",
"source_idx": 4612,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1575,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{2008}-2^{2008}",
"budget": 1260
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004612",
"normalized_example_id": "ex_004612",
"source_idx": 4612,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1575,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{2008}-2^{2008}",
"budget": 1575
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004612",
"normalized_example_id": "ex_004612",
"source_idx": 4612,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1575,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{6075}{32}",
"budget": 3389
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004613",
"normalized_example_id": "ex_004613",
"source_idx": 4613,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5649,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{6075}{32}",
"budget": 4519
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004613",
"normalized_example_id": "ex_004613",
"source_idx": 4613,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5649,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{6075}{32}",
"budget": 5649
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004613",
"normalized_example_id": "ex_004613",
"source_idx": 4613,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5649,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "17^{14}",
"budget": 4919
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004614",
"normalized_example_id": "ex_004614",
"source_idx": 4614,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8198,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "17^{14}",
"budget": 6558
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004614",
"normalized_example_id": "ex_004614",
"source_idx": 4614,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8198,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "17^{14}",
"budget": 8198
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004614",
"normalized_example_id": "ex_004614",
"source_idx": 4614,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8198,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "26",
"budget": 790
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004615",
"normalized_example_id": "ex_004615",
"source_idx": 4615,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1317,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "26",
"budget": 1054
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004615",
"normalized_example_id": "ex_004615",
"source_idx": 4615,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1317,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "26",
"budget": 1317
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004615",
"normalized_example_id": "ex_004615",
"source_idx": 4615,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1317,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4864
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004616",
"normalized_example_id": "ex_004616",
"source_idx": 4616,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8107,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6486
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004616",
"normalized_example_id": "ex_004616",
"source_idx": 4616,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8107,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 8107
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004616",
"normalized_example_id": "ex_004616",
"source_idx": 4616,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8107,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}=n",
"budget": 2161
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004620",
"normalized_example_id": "ex_004620",
"source_idx": 4620,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3602,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}=n",
"budget": 2882
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004620",
"normalized_example_id": "ex_004620",
"source_idx": 4620,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3602,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}=n",
"budget": 3602
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004620",
"normalized_example_id": "ex_004620",
"source_idx": 4620,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3602,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{41\\pi\\sqrt{41}}{384}",
"budget": 4862
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004621",
"normalized_example_id": "ex_004621",
"source_idx": 4621,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8104,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{41\\pi\\sqrt{41}}{384}",
"budget": 6483
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004621",
"normalized_example_id": "ex_004621",
"source_idx": 4621,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8104,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{41\\pi\\sqrt{41}}{384}",
"budget": 8104
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004621",
"normalized_example_id": "ex_004621",
"source_idx": 4621,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8104,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "82.2",
"budget": 3319
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004623",
"normalized_example_id": "ex_004623",
"source_idx": 4623,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5532,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "82.2",
"budget": 4426
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004623",
"normalized_example_id": "ex_004623",
"source_idx": 4623,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5532,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "82.2",
"budget": 5532
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004623",
"normalized_example_id": "ex_004623",
"source_idx": 4623,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5532,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3537",
"budget": 2047
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004631",
"normalized_example_id": "ex_004631",
"source_idx": 4631,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3411,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3537",
"budget": 2729
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004631",
"normalized_example_id": "ex_004631",
"source_idx": 4631,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3411,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3537",
"budget": 3411
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004631",
"normalized_example_id": "ex_004631",
"source_idx": 4631,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3411,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2271
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004633",
"normalized_example_id": "ex_004633",
"source_idx": 4633,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3785,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3028
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004633",
"normalized_example_id": "ex_004633",
"source_idx": 4633,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3785,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3785
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004633",
"normalized_example_id": "ex_004633",
"source_idx": 4633,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3785,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 3484
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004634",
"normalized_example_id": "ex_004634",
"source_idx": 4634,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5807,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 4646
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004634",
"normalized_example_id": "ex_004634",
"source_idx": 4634,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5807,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 5807
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004634",
"normalized_example_id": "ex_004634",
"source_idx": 4634,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5807,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{120}{169}",
"budget": 1579
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004636",
"normalized_example_id": "ex_004636",
"source_idx": 4636,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2632,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{120}{169}",
"budget": 2106
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004636",
"normalized_example_id": "ex_004636",
"source_idx": 4636,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2632,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{120}{169}",
"budget": 2632
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004636",
"normalized_example_id": "ex_004636",
"source_idx": 4636,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2632,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{1}=0.5,y_{1}=-0.5;x_{2}=0,y_{2}=0;x_{3}=1,y_{3}=0;x_{4}=0.5,y_{4}=0.5",
"budget": 2132
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004639",
"normalized_example_id": "ex_004639",
"source_idx": 4639,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3553,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{1}=0.5,y_{1}=-0.5;x_{2}=0,y_{2}=0;x_{3}=1,y_{3}=0;x_{4}=0.5,y_{4}=0.5",
"budget": 2842
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004639",
"normalized_example_id": "ex_004639",
"source_idx": 4639,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3553,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{1}=0.5,y_{1}=-0.5;x_{2}=0,y_{2}=0;x_{3}=1,y_{3}=0;x_{4}=0.5,y_{4}=0.5",
"budget": 3553
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004639",
"normalized_example_id": "ex_004639",
"source_idx": 4639,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3553,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1+3\\cos2\\alpha}{4}",
"budget": 5290
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004641",
"normalized_example_id": "ex_004641",
"source_idx": 4641,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8816,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1+3\\cos2\\alpha}{4}",
"budget": 7053
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004641",
"normalized_example_id": "ex_004641",
"source_idx": 4641,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8816,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1+3\\cos2\\alpha}{4}",
"budget": 8816
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004641",
"normalized_example_id": "ex_004641",
"source_idx": 4641,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8816,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(n, k) = (5, 2)",
"budget": 5569
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004642",
"normalized_example_id": "ex_004642",
"source_idx": 4642,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9282,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(n, k) = (5, 2)",
"budget": 7426
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004642",
"normalized_example_id": "ex_004642",
"source_idx": 4642,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9282,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(n, k) = (5, 2)",
"budget": 9282
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004642",
"normalized_example_id": "ex_004642",
"source_idx": 4642,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9282,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "507",
"budget": 2170
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004643",
"normalized_example_id": "ex_004643",
"source_idx": 4643,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3616,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "507",
"budget": 2893
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004643",
"normalized_example_id": "ex_004643",
"source_idx": 4643,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3616,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "507",
"budget": 3616
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004643",
"normalized_example_id": "ex_004643",
"source_idx": 4643,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3616,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "43",
"budget": 1932
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004645",
"normalized_example_id": "ex_004645",
"source_idx": 4645,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3220,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "43",
"budget": 2576
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004645",
"normalized_example_id": "ex_004645",
"source_idx": 4645,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3220,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "43",
"budget": 3220
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004645",
"normalized_example_id": "ex_004645",
"source_idx": 4645,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3220,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2015",
"budget": 4234
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004646",
"normalized_example_id": "ex_004646",
"source_idx": 4646,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7057,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2015",
"budget": 5646
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004646",
"normalized_example_id": "ex_004646",
"source_idx": 4646,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7057,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2015",
"budget": 7057
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004646",
"normalized_example_id": "ex_004646",
"source_idx": 4646,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7057,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1568
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004647",
"normalized_example_id": "ex_004647",
"source_idx": 4647,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2091
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004647",
"normalized_example_id": "ex_004647",
"source_idx": 4647,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2614
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004647",
"normalized_example_id": "ex_004647",
"source_idx": 4647,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "81,192,399,501,1008,2019",
"budget": 2717
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004649",
"normalized_example_id": "ex_004649",
"source_idx": 4649,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4528,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "81,192,399,501,1008,2019",
"budget": 3622
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004649",
"normalized_example_id": "ex_004649",
"source_idx": 4649,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4528,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "81,192,399,501,1008,2019",
"budget": 4528
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004649",
"normalized_example_id": "ex_004649",
"source_idx": 4649,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4528,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1303
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004653",
"normalized_example_id": "ex_004653",
"source_idx": 4653,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2171,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1737
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004653",
"normalized_example_id": "ex_004653",
"source_idx": 4653,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2171,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2171
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004653",
"normalized_example_id": "ex_004653",
"source_idx": 4653,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2171,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 953
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004654",
"normalized_example_id": "ex_004654",
"source_idx": 4654,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1589,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1271
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004654",
"normalized_example_id": "ex_004654",
"source_idx": 4654,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1589,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1589
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004654",
"normalized_example_id": "ex_004654",
"source_idx": 4654,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1589,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3640",
"budget": 1477
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004655",
"normalized_example_id": "ex_004655",
"source_idx": 4655,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2462,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3640",
"budget": 1970
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004655",
"normalized_example_id": "ex_004655",
"source_idx": 4655,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2462,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3640",
"budget": 2462
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004655",
"normalized_example_id": "ex_004655",
"source_idx": 4655,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2462,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+\\frac{1}{2}\\ln\\frac{3}{2}",
"budget": 5841
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004656",
"normalized_example_id": "ex_004656",
"source_idx": 4656,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9735,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+\\frac{1}{2}\\ln\\frac{3}{2}",
"budget": 7788
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004656",
"normalized_example_id": "ex_004656",
"source_idx": 4656,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9735,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+\\frac{1}{2}\\ln\\frac{3}{2}",
"budget": 9735
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004656",
"normalized_example_id": "ex_004656",
"source_idx": 4656,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9735,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.