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": "\\frac{1}{\\sqrt{8}}",
"budget": 2544
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004864",
"normalized_example_id": "ex_004864",
"source_idx": 4864,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4240,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{\\sqrt{8}}",
"budget": 3392
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004864",
"normalized_example_id": "ex_004864",
"source_idx": 4864,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4240,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{\\sqrt{8}}",
"budget": 4240
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004864",
"normalized_example_id": "ex_004864",
"source_idx": 4864,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4240,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1142
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004868",
"normalized_example_id": "ex_004868",
"source_idx": 4868,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1903,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1522
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004868",
"normalized_example_id": "ex_004868",
"source_idx": 4868,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1903,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1903
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004868",
"normalized_example_id": "ex_004868",
"source_idx": 4868,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1903,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{55}{96}",
"budget": 4313
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004869",
"normalized_example_id": "ex_004869",
"source_idx": 4869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7188,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{55}{96}",
"budget": 5750
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004869",
"normalized_example_id": "ex_004869",
"source_idx": 4869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7188,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{55}{96}",
"budget": 7188
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004869",
"normalized_example_id": "ex_004869",
"source_idx": 4869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7188,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 2454
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004870",
"normalized_example_id": "ex_004870",
"source_idx": 4870,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4090,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 3272
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004870",
"normalized_example_id": "ex_004870",
"source_idx": 4870,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4090,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 4090
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004870",
"normalized_example_id": "ex_004870",
"source_idx": 4870,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4090,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 3311
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004871",
"normalized_example_id": "ex_004871",
"source_idx": 4871,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5518,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 4414
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004871",
"normalized_example_id": "ex_004871",
"source_idx": 4871,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5518,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 5518
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004871",
"normalized_example_id": "ex_004871",
"source_idx": 4871,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5518,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[\\frac{3}{4},+\\infty)",
"budget": 2179
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004875",
"normalized_example_id": "ex_004875",
"source_idx": 4875,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3631,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[\\frac{3}{4},+\\infty)",
"budget": 2905
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004875",
"normalized_example_id": "ex_004875",
"source_idx": 4875,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3631,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[\\frac{3}{4},+\\infty)",
"budget": 3631
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004875",
"normalized_example_id": "ex_004875",
"source_idx": 4875,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3631,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6a",
"budget": 2132
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004876",
"normalized_example_id": "ex_004876",
"source_idx": 4876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3554,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6a",
"budget": 2843
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004876",
"normalized_example_id": "ex_004876",
"source_idx": 4876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3554,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6a",
"budget": 3554
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004876",
"normalized_example_id": "ex_004876",
"source_idx": 4876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3554,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "140",
"budget": 1426
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004877",
"normalized_example_id": "ex_004877",
"source_idx": 4877,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "140",
"budget": 1902
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004877",
"normalized_example_id": "ex_004877",
"source_idx": 4877,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "140",
"budget": 2377
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004877",
"normalized_example_id": "ex_004877",
"source_idx": 4877,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 3223
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004881",
"normalized_example_id": "ex_004881",
"source_idx": 4881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5371,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 4297
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004881",
"normalized_example_id": "ex_004881",
"source_idx": 4881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5371,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 5371
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004881",
"normalized_example_id": "ex_004881",
"source_idx": 4881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5371,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 3923
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004882",
"normalized_example_id": "ex_004882",
"source_idx": 4882,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6539,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 5231
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004882",
"normalized_example_id": "ex_004882",
"source_idx": 4882,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6539,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 6539
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004882",
"normalized_example_id": "ex_004882",
"source_idx": 4882,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6539,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(4k,(-\\frac{1}{4})^k)",
"budget": 2327
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004883",
"normalized_example_id": "ex_004883",
"source_idx": 4883,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3878,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(4k,(-\\frac{1}{4})^k)",
"budget": 3102
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004883",
"normalized_example_id": "ex_004883",
"source_idx": 4883,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3878,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(4k,(-\\frac{1}{4})^k)",
"budget": 3878
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004883",
"normalized_example_id": "ex_004883",
"source_idx": 4883,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3878,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "840",
"budget": 6043
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004884",
"normalized_example_id": "ex_004884",
"source_idx": 4884,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10072,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "840",
"budget": 8058
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004884",
"normalized_example_id": "ex_004884",
"source_idx": 4884,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10072,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "840",
"budget": 10072
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004884",
"normalized_example_id": "ex_004884",
"source_idx": 4884,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10072,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\pi",
"budget": 1596
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004885",
"normalized_example_id": "ex_004885",
"source_idx": 4885,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2660,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\pi",
"budget": 2128
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004885",
"normalized_example_id": "ex_004885",
"source_idx": 4885,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2660,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\pi",
"budget": 2660
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004885",
"normalized_example_id": "ex_004885",
"source_idx": 4885,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2660,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "130051",
"budget": 4232
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004888",
"normalized_example_id": "ex_004888",
"source_idx": 4888,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7053,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "130051",
"budget": 5642
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004888",
"normalized_example_id": "ex_004888",
"source_idx": 4888,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7053,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "130051",
"budget": 7053
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004888",
"normalized_example_id": "ex_004888",
"source_idx": 4888,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7053,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 3419
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004889",
"normalized_example_id": "ex_004889",
"source_idx": 4889,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5699,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 4559
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004889",
"normalized_example_id": "ex_004889",
"source_idx": 4889,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5699,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 5699
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004889",
"normalized_example_id": "ex_004889",
"source_idx": 4889,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5699,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 4963
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004890",
"normalized_example_id": "ex_004890",
"source_idx": 4890,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8271,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 6617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004890",
"normalized_example_id": "ex_004890",
"source_idx": 4890,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8271,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 8271
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004890",
"normalized_example_id": "ex_004890",
"source_idx": 4890,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8271,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 792
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004894",
"normalized_example_id": "ex_004894",
"source_idx": 4894,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1320,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 1056
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004894",
"normalized_example_id": "ex_004894",
"source_idx": 4894,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1320,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 1320
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004894",
"normalized_example_id": "ex_004894",
"source_idx": 4894,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1320,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "63",
"budget": 2261
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004895",
"normalized_example_id": "ex_004895",
"source_idx": 4895,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3768,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "63",
"budget": 3014
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004895",
"normalized_example_id": "ex_004895",
"source_idx": 4895,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3768,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "63",
"budget": 3768
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004895",
"normalized_example_id": "ex_004895",
"source_idx": 4895,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3768,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[2,3]",
"budget": 3700
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004898",
"normalized_example_id": "ex_004898",
"source_idx": 4898,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6166,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[2,3]",
"budget": 4933
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004898",
"normalized_example_id": "ex_004898",
"source_idx": 4898,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6166,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[2,3]",
"budget": 6166
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004898",
"normalized_example_id": "ex_004898",
"source_idx": 4898,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6166,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3}{16}",
"budget": 3990
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004900",
"normalized_example_id": "ex_004900",
"source_idx": 4900,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6650,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3}{16}",
"budget": 5320
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004900",
"normalized_example_id": "ex_004900",
"source_idx": 4900,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6650,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3}{16}",
"budget": 6650
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004900",
"normalized_example_id": "ex_004900",
"source_idx": 4900,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6650,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "21",
"budget": 5629
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004901",
"normalized_example_id": "ex_004901",
"source_idx": 4901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9381,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "21",
"budget": 7505
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004901",
"normalized_example_id": "ex_004901",
"source_idx": 4901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9381,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "21",
"budget": 9381
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004901",
"normalized_example_id": "ex_004901",
"source_idx": 4901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9381,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-34",
"budget": 3557
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004902",
"normalized_example_id": "ex_004902",
"source_idx": 4902,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5928,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-34",
"budget": 4742
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004902",
"normalized_example_id": "ex_004902",
"source_idx": 4902,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5928,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-34",
"budget": 5928
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004902",
"normalized_example_id": "ex_004902",
"source_idx": 4902,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5928,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "k = \\left\\lceil \\frac{n}{2} \\right\\rceil + 1",
"budget": 4398
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004906",
"normalized_example_id": "ex_004906",
"source_idx": 4906,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7330,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "k = \\left\\lceil \\frac{n}{2} \\right\\rceil + 1",
"budget": 5864
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004906",
"normalized_example_id": "ex_004906",
"source_idx": 4906,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7330,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "k = \\left\\lceil \\frac{n}{2} \\right\\rceil + 1",
"budget": 7330
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004906",
"normalized_example_id": "ex_004906",
"source_idx": 4906,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7330,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{22}{21}",
"budget": 1871
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004907",
"normalized_example_id": "ex_004907",
"source_idx": 4907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3118,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{22}{21}",
"budget": 2494
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004907",
"normalized_example_id": "ex_004907",
"source_idx": 4907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3118,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{22}{21}",
"budget": 3118
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004907",
"normalized_example_id": "ex_004907",
"source_idx": 4907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3118,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{n-1}",
"budget": 2675
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004908",
"normalized_example_id": "ex_004908",
"source_idx": 4908,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4458,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{n-1}",
"budget": 3566
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004908",
"normalized_example_id": "ex_004908",
"source_idx": 4908,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4458,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{n-1}",
"budget": 4458
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004908",
"normalized_example_id": "ex_004908",
"source_idx": 4908,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4458,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1011,1012,1013,1015",
"budget": 5240
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004909",
"normalized_example_id": "ex_004909",
"source_idx": 4909,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8733,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1011,1012,1013,1015",
"budget": 6986
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004909",
"normalized_example_id": "ex_004909",
"source_idx": 4909,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8733,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1011,1012,1013,1015",
"budget": 8733
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004909",
"normalized_example_id": "ex_004909",
"source_idx": 4909,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8733,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3689
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004910",
"normalized_example_id": "ex_004910",
"source_idx": 4910,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6149,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4919
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004910",
"normalized_example_id": "ex_004910",
"source_idx": 4910,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6149,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6149
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004910",
"normalized_example_id": "ex_004910",
"source_idx": 4910,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6149,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 1621
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004914",
"normalized_example_id": "ex_004914",
"source_idx": 4914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2702,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 2162
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004914",
"normalized_example_id": "ex_004914",
"source_idx": 4914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2702,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 2702
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004914",
"normalized_example_id": "ex_004914",
"source_idx": 4914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2702,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7290",
"budget": 2445
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004915",
"normalized_example_id": "ex_004915",
"source_idx": 4915,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4075,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7290",
"budget": 3260
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004915",
"normalized_example_id": "ex_004915",
"source_idx": 4915,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4075,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7290",
"budget": 4075
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004915",
"normalized_example_id": "ex_004915",
"source_idx": 4915,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4075,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3}{4}",
"budget": 1075
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004916",
"normalized_example_id": "ex_004916",
"source_idx": 4916,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1792,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3}{4}",
"budget": 1434
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004916",
"normalized_example_id": "ex_004916",
"source_idx": 4916,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1792,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3}{4}",
"budget": 1792
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004916",
"normalized_example_id": "ex_004916",
"source_idx": 4916,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1792,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 707
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004917",
"normalized_example_id": "ex_004917",
"source_idx": 4917,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1179,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 943
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004917",
"normalized_example_id": "ex_004917",
"source_idx": 4917,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1179,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 1179
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004917",
"normalized_example_id": "ex_004917",
"source_idx": 4917,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1179,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+6\\sqrt{7}",
"budget": 2822
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004918",
"normalized_example_id": "ex_004918",
"source_idx": 4918,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4704,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+6\\sqrt{7}",
"budget": 3763
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004918",
"normalized_example_id": "ex_004918",
"source_idx": 4918,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4704,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+6\\sqrt{7}",
"budget": 4704
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004918",
"normalized_example_id": "ex_004918",
"source_idx": 4918,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4704,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "34",
"budget": 1480
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004920",
"normalized_example_id": "ex_004920",
"source_idx": 4920,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "34",
"budget": 1974
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004920",
"normalized_example_id": "ex_004920",
"source_idx": 4920,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "34",
"budget": 2467
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004920",
"normalized_example_id": "ex_004920",
"source_idx": 4920,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{7}",
"budget": 3373
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004921",
"normalized_example_id": "ex_004921",
"source_idx": 4921,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5621,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.