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": "53",
"budget": 2617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003433",
"normalized_example_id": "ex_003433",
"source_idx": 3433,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4361,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "53",
"budget": 3489
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003433",
"normalized_example_id": "ex_003433",
"source_idx": 3433,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4361,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "53",
"budget": 4361
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003433",
"normalized_example_id": "ex_003433",
"source_idx": 3433,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4361,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{x}",
"budget": 1291
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003436",
"normalized_example_id": "ex_003436",
"source_idx": 3436,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2152,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{x}",
"budget": 1722
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003436",
"normalized_example_id": "ex_003436",
"source_idx": 3436,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2152,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{x}",
"budget": 2152
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003436",
"normalized_example_id": "ex_003436",
"source_idx": 3436,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2152,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4118
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003438",
"normalized_example_id": "ex_003438",
"source_idx": 3438,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6863,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5490
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003438",
"normalized_example_id": "ex_003438",
"source_idx": 3438,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6863,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6863
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003438",
"normalized_example_id": "ex_003438",
"source_idx": 3438,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6863,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "44",
"budget": 1727
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003440",
"normalized_example_id": "ex_003440",
"source_idx": 3440,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2878,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "44",
"budget": 2302
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003440",
"normalized_example_id": "ex_003440",
"source_idx": 3440,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2878,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "44",
"budget": 2878
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003440",
"normalized_example_id": "ex_003440",
"source_idx": 3440,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2878,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 2026
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003441",
"normalized_example_id": "ex_003441",
"source_idx": 3441,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 2702
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003441",
"normalized_example_id": "ex_003441",
"source_idx": 3441,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 3377
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003441",
"normalized_example_id": "ex_003441",
"source_idx": 3441,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\ln^{2}3}",
"budget": 2162
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003442",
"normalized_example_id": "ex_003442",
"source_idx": 3442,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3604,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\ln^{2}3}",
"budget": 2883
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003442",
"normalized_example_id": "ex_003442",
"source_idx": 3442,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3604,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\ln^{2}3}",
"budget": 3604
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003442",
"normalized_example_id": "ex_003442",
"source_idx": 3442,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3604,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 935
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003446",
"normalized_example_id": "ex_003446",
"source_idx": 3446,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1558,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1246
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003446",
"normalized_example_id": "ex_003446",
"source_idx": 3446,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1558,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1558
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003446",
"normalized_example_id": "ex_003446",
"source_idx": 3446,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1558,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "29",
"budget": 3061
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003448",
"normalized_example_id": "ex_003448",
"source_idx": 3448,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5102,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "29",
"budget": 4082
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003448",
"normalized_example_id": "ex_003448",
"source_idx": 3448,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5102,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "29",
"budget": 5102
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003448",
"normalized_example_id": "ex_003448",
"source_idx": 3448,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5102,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18:7",
"budget": 2347
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003451",
"normalized_example_id": "ex_003451",
"source_idx": 3451,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3912,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18:7",
"budget": 3130
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003451",
"normalized_example_id": "ex_003451",
"source_idx": 3451,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3912,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18:7",
"budget": 3912
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003451",
"normalized_example_id": "ex_003451",
"source_idx": 3451,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3912,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(1,1,1)",
"budget": 2855
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003456",
"normalized_example_id": "ex_003456",
"source_idx": 3456,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4758,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(1,1,1)",
"budget": 3806
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003456",
"normalized_example_id": "ex_003456",
"source_idx": 3456,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4758,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(1,1,1)",
"budget": 4758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003456",
"normalized_example_id": "ex_003456",
"source_idx": 3456,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4758,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "28",
"budget": 3052
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003457",
"normalized_example_id": "ex_003457",
"source_idx": 3457,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5086,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "28",
"budget": 4069
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003457",
"normalized_example_id": "ex_003457",
"source_idx": 3457,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5086,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "28",
"budget": 5086
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003457",
"normalized_example_id": "ex_003457",
"source_idx": 3457,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5086,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(1,2),(2,1),(-1,-2),(-2,-1)",
"budget": 3274
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003459",
"normalized_example_id": "ex_003459",
"source_idx": 3459,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5456,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(1,2),(2,1),(-1,-2),(-2,-1)",
"budget": 4365
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003459",
"normalized_example_id": "ex_003459",
"source_idx": 3459,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5456,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(1,2),(2,1),(-1,-2),(-2,-1)",
"budget": 5456
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003459",
"normalized_example_id": "ex_003459",
"source_idx": 3459,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5456,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1762
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003461",
"normalized_example_id": "ex_003461",
"source_idx": 3461,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2936,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2349
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003461",
"normalized_example_id": "ex_003461",
"source_idx": 3461,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2936,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2936
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003461",
"normalized_example_id": "ex_003461",
"source_idx": 3461,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2936,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2\\sqrt{3}}{3}",
"budget": 2938
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003463",
"normalized_example_id": "ex_003463",
"source_idx": 3463,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4896,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2\\sqrt{3}}{3}",
"budget": 3917
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003463",
"normalized_example_id": "ex_003463",
"source_idx": 3463,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4896,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2\\sqrt{3}}{3}",
"budget": 4896
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003463",
"normalized_example_id": "ex_003463",
"source_idx": 3463,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4896,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\epsilon_0 L^3 \\theta \\xi}{d^2}",
"budget": 4402
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003464",
"normalized_example_id": "ex_003464",
"source_idx": 3464,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7337,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\epsilon_0 L^3 \\theta \\xi}{d^2}",
"budget": 5870
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003464",
"normalized_example_id": "ex_003464",
"source_idx": 3464,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7337,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{\\epsilon_0 L^3 \\theta \\xi}{d^2}",
"budget": 7337
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003464",
"normalized_example_id": "ex_003464",
"source_idx": 3464,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7337,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "62",
"budget": 820
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003466",
"normalized_example_id": "ex_003466",
"source_idx": 3466,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1367,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "62",
"budget": 1094
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003466",
"normalized_example_id": "ex_003466",
"source_idx": 3466,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1367,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "62",
"budget": 1367
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003466",
"normalized_example_id": "ex_003466",
"source_idx": 3466,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1367,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1748
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003468",
"normalized_example_id": "ex_003468",
"source_idx": 3468,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2914,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 2331
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003468",
"normalized_example_id": "ex_003468",
"source_idx": 3468,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2914,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 2914
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003468",
"normalized_example_id": "ex_003468",
"source_idx": 3468,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2914,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-3,2,5",
"budget": 5898
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003470",
"normalized_example_id": "ex_003470",
"source_idx": 3470,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9830,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-3,2,5",
"budget": 7864
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003470",
"normalized_example_id": "ex_003470",
"source_idx": 3470,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9830,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-3,2,5",
"budget": 9830
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003470",
"normalized_example_id": "ex_003470",
"source_idx": 3470,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9830,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;6;-27),(6;1;-27),(1;7;63),(7;1;63)",
"budget": 2435
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003471",
"normalized_example_id": "ex_003471",
"source_idx": 3471,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4058,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;6;-27),(6;1;-27),(1;7;63),(7;1;63)",
"budget": 3246
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003471",
"normalized_example_id": "ex_003471",
"source_idx": 3471,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4058,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;6;-27),(6;1;-27),(1;7;63),(7;1;63)",
"budget": 4058
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003471",
"normalized_example_id": "ex_003471",
"source_idx": 3471,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4058,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2017",
"budget": 4222
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003473",
"normalized_example_id": "ex_003473",
"source_idx": 3473,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7036,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2017",
"budget": 5629
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003473",
"normalized_example_id": "ex_003473",
"source_idx": 3473,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7036,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2017",
"budget": 7036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003473",
"normalized_example_id": "ex_003473",
"source_idx": 3473,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7036,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9or31",
"budget": 3301
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003475",
"normalized_example_id": "ex_003475",
"source_idx": 3475,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5501,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9or31",
"budget": 4401
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003475",
"normalized_example_id": "ex_003475",
"source_idx": 3475,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5501,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9or31",
"budget": 5501
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003475",
"normalized_example_id": "ex_003475",
"source_idx": 3475,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5501,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{n}=\\frac{3^{n}-1}{2}X+\\frac{3-3^{n}}{2}",
"budget": 2750
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003476",
"normalized_example_id": "ex_003476",
"source_idx": 3476,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4584,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{n}=\\frac{3^{n}-1}{2}X+\\frac{3-3^{n}}{2}",
"budget": 3667
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003476",
"normalized_example_id": "ex_003476",
"source_idx": 3476,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4584,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{n}=\\frac{3^{n}-1}{2}X+\\frac{3-3^{n}}{2}",
"budget": 4584
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003476",
"normalized_example_id": "ex_003476",
"source_idx": 3476,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4584,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1171
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003477",
"normalized_example_id": "ex_003477",
"source_idx": 3477,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1951,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1561
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003477",
"normalized_example_id": "ex_003477",
"source_idx": 3477,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1951,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1951
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003477",
"normalized_example_id": "ex_003477",
"source_idx": 3477,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1951,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "120",
"budget": 1036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003480",
"normalized_example_id": "ex_003480",
"source_idx": 3480,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1727,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "120",
"budget": 1382
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003480",
"normalized_example_id": "ex_003480",
"source_idx": 3480,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1727,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "120",
"budget": 1727
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003480",
"normalized_example_id": "ex_003480",
"source_idx": 3480,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1727,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1023
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003482",
"normalized_example_id": "ex_003482",
"source_idx": 3482,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1364
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003482",
"normalized_example_id": "ex_003482",
"source_idx": 3482,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1705
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003482",
"normalized_example_id": "ex_003482",
"source_idx": 3482,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "627",
"budget": 3283
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003487",
"normalized_example_id": "ex_003487",
"source_idx": 3487,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5472,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "627",
"budget": 4378
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003487",
"normalized_example_id": "ex_003487",
"source_idx": 3487,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5472,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "627",
"budget": 5472
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003487",
"normalized_example_id": "ex_003487",
"source_idx": 3487,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5472,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2}{3}",
"budget": 2449
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003490",
"normalized_example_id": "ex_003490",
"source_idx": 3490,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4081,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2}{3}",
"budget": 3265
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003490",
"normalized_example_id": "ex_003490",
"source_idx": 3490,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4081,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2}{3}",
"budget": 4081
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003490",
"normalized_example_id": "ex_003490",
"source_idx": 3490,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4081,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;4;-83),(4;1;-83),(1;5;-79),(5;1;-79)",
"budget": 5731
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003493",
"normalized_example_id": "ex_003493",
"source_idx": 3493,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9552,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;4;-83),(4;1;-83),(1;5;-79),(5;1;-79)",
"budget": 7642
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003493",
"normalized_example_id": "ex_003493",
"source_idx": 3493,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9552,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;4;-83),(4;1;-83),(1;5;-79),(5;1;-79)",
"budget": 9552
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003493",
"normalized_example_id": "ex_003493",
"source_idx": 3493,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9552,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\equiv 0, \\pm 1(\\bmod 9)",
"budget": 1387
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003496",
"normalized_example_id": "ex_003496",
"source_idx": 3496,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2312,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\equiv 0, \\pm 1(\\bmod 9)",
"budget": 1850
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003496",
"normalized_example_id": "ex_003496",
"source_idx": 3496,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2312,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\equiv 0, \\pm 1(\\bmod 9)",
"budget": 2312
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003496",
"normalized_example_id": "ex_003496",
"source_idx": 3496,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2312,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x)=\\lfloorx\\rfloor",
"budget": 5670
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003497",
"normalized_example_id": "ex_003497",
"source_idx": 3497,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9450,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x)=\\lfloorx\\rfloor",
"budget": 7560
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003497",
"normalized_example_id": "ex_003497",
"source_idx": 3497,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9450,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x)=\\lfloorx\\rfloor",
"budget": 9450
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003497",
"normalized_example_id": "ex_003497",
"source_idx": 3497,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9450,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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=-\\frac{3}{2} \\sqrt{5}",
"budget": 2990
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003498",
"normalized_example_id": "ex_003498",
"source_idx": 3498,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4983,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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=-\\frac{3}{2} \\sqrt{5}",
"budget": 3986
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003498",
"normalized_example_id": "ex_003498",
"source_idx": 3498,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4983,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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=-\\frac{3}{2} \\sqrt{5}",
"budget": 4983
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003498",
"normalized_example_id": "ex_003498",
"source_idx": 3498,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4983,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{2}+1)(y^{2}-1)=C",
"budget": 2905
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003500",
"normalized_example_id": "ex_003500",
"source_idx": 3500,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4842,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{2}+1)(y^{2}-1)=C",
"budget": 3874
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003500",
"normalized_example_id": "ex_003500",
"source_idx": 3500,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4842,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^{2}+1)(y^{2}-1)=C",
"budget": 4842
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003500",
"normalized_example_id": "ex_003500",
"source_idx": 3500,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4842,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3",
"budget": 4861
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003506",
"normalized_example_id": "ex_003506",
"source_idx": 3506,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8102,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3",
"budget": 6482
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003506",
"normalized_example_id": "ex_003506",
"source_idx": 3506,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8102,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3",
"budget": 8102
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003506",
"normalized_example_id": "ex_003506",
"source_idx": 3506,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8102,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0.178",
"budget": 1671
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003507",
"normalized_example_id": "ex_003507",
"source_idx": 3507,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2785,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.