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": "7",
"budget": 2920
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004659",
"normalized_example_id": "ex_004659",
"source_idx": 4659,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4866,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3893
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004659",
"normalized_example_id": "ex_004659",
"source_idx": 4659,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4866,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4866
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004659",
"normalized_example_id": "ex_004659",
"source_idx": 4659,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4866,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "15",
"budget": 2504
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004660",
"normalized_example_id": "ex_004660",
"source_idx": 4660,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4173,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "15",
"budget": 3338
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004660",
"normalized_example_id": "ex_004660",
"source_idx": 4660,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4173,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "15",
"budget": 4173
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004660",
"normalized_example_id": "ex_004660",
"source_idx": 4660,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4173,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "y = \\frac{3x^2}{4}",
"budget": 5646
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004667",
"normalized_example_id": "ex_004667",
"source_idx": 4667,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9410,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "y = \\frac{3x^2}{4}",
"budget": 7528
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004667",
"normalized_example_id": "ex_004667",
"source_idx": 4667,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9410,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "y = \\frac{3x^2}{4}",
"budget": 9410
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004667",
"normalized_example_id": "ex_004667",
"source_idx": 4667,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9410,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2418
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004669",
"normalized_example_id": "ex_004669",
"source_idx": 4669,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4030,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3224
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004669",
"normalized_example_id": "ex_004669",
"source_idx": 4669,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4030,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4030
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004669",
"normalized_example_id": "ex_004669",
"source_idx": 4669,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4030,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35,52",
"budget": 4467
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004670",
"normalized_example_id": "ex_004670",
"source_idx": 4670,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7445,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35,52",
"budget": 5956
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004670",
"normalized_example_id": "ex_004670",
"source_idx": 4670,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7445,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35,52",
"budget": 7445
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004670",
"normalized_example_id": "ex_004670",
"source_idx": 4670,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7445,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7,14,21,28,35",
"budget": 1192
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004672",
"normalized_example_id": "ex_004672",
"source_idx": 4672,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1986,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7,14,21,28,35",
"budget": 1589
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004672",
"normalized_example_id": "ex_004672",
"source_idx": 4672,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1986,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7,14,21,28,35",
"budget": 1986
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004672",
"normalized_example_id": "ex_004672",
"source_idx": 4672,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1986,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "23",
"budget": 1732
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004674",
"normalized_example_id": "ex_004674",
"source_idx": 4674,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2886,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "23",
"budget": 2309
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004674",
"normalized_example_id": "ex_004674",
"source_idx": 4674,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2886,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "23",
"budget": 2886
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004674",
"normalized_example_id": "ex_004674",
"source_idx": 4674,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2886,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "216",
"budget": 758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004675",
"normalized_example_id": "ex_004675",
"source_idx": 4675,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1264,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "216",
"budget": 1011
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004675",
"normalized_example_id": "ex_004675",
"source_idx": 4675,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1264,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "216",
"budget": 1264
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004675",
"normalized_example_id": "ex_004675",
"source_idx": 4675,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1264,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "666",
"budget": 3181
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004677",
"normalized_example_id": "ex_004677",
"source_idx": 4677,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5302,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "666",
"budget": 4242
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004677",
"normalized_example_id": "ex_004677",
"source_idx": 4677,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5302,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "666",
"budget": 5302
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004677",
"normalized_example_id": "ex_004677",
"source_idx": 4677,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5302,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "414\\cdot9!",
"budget": 3995
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004680",
"normalized_example_id": "ex_004680",
"source_idx": 4680,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6658,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "414\\cdot9!",
"budget": 5326
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004680",
"normalized_example_id": "ex_004680",
"source_idx": 4680,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6658,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "414\\cdot9!",
"budget": 6658
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004680",
"normalized_example_id": "ex_004680",
"source_idx": 4680,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6658,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{a b c}{\\sqrt{a^{2} b^{2}+b^{2} c^{2}+a^{2} c^{2}}}",
"budget": 3322
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004682",
"normalized_example_id": "ex_004682",
"source_idx": 4682,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5537,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{a b c}{\\sqrt{a^{2} b^{2}+b^{2} c^{2}+a^{2} c^{2}}}",
"budget": 4430
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004682",
"normalized_example_id": "ex_004682",
"source_idx": 4682,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5537,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{a b c}{\\sqrt{a^{2} b^{2}+b^{2} c^{2}+a^{2} c^{2}}}",
"budget": 5537
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004682",
"normalized_example_id": "ex_004682",
"source_idx": 4682,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5537,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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)=0 \\text{ or } f(x)=x^2",
"budget": 3502
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004683",
"normalized_example_id": "ex_004683",
"source_idx": 4683,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5837,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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)=0 \\text{ or } f(x)=x^2",
"budget": 4670
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004683",
"normalized_example_id": "ex_004683",
"source_idx": 4683,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5837,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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)=0 \\text{ or } f(x)=x^2",
"budget": 5837
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004683",
"normalized_example_id": "ex_004683",
"source_idx": 4683,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5837,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "24\\frac{1}{5}",
"budget": 2809
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004684",
"normalized_example_id": "ex_004684",
"source_idx": 4684,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4682,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "24\\frac{1}{5}",
"budget": 3746
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004684",
"normalized_example_id": "ex_004684",
"source_idx": 4684,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4682,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "24\\frac{1}{5}",
"budget": 4682
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004684",
"normalized_example_id": "ex_004684",
"source_idx": 4684,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4682,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 4328
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004685",
"normalized_example_id": "ex_004685",
"source_idx": 4685,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7213,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 5770
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004685",
"normalized_example_id": "ex_004685",
"source_idx": 4685,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7213,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 7213
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004685",
"normalized_example_id": "ex_004685",
"source_idx": 4685,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7213,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\sqrt{2}-2",
"budget": 1805
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004687",
"normalized_example_id": "ex_004687",
"source_idx": 4687,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3009,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\sqrt{2}-2",
"budget": 2407
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004687",
"normalized_example_id": "ex_004687",
"source_idx": 4687,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3009,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\sqrt{2}-2",
"budget": 3009
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004687",
"normalized_example_id": "ex_004687",
"source_idx": 4687,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3009,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "36,6,12,33",
"budget": 1432
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004688",
"normalized_example_id": "ex_004688",
"source_idx": 4688,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2387,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "36,6,12,33",
"budget": 1910
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004688",
"normalized_example_id": "ex_004688",
"source_idx": 4688,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2387,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "36,6,12,33",
"budget": 2387
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004688",
"normalized_example_id": "ex_004688",
"source_idx": 4688,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2387,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4243
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004692",
"normalized_example_id": "ex_004692",
"source_idx": 4692,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7071,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5657
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004692",
"normalized_example_id": "ex_004692",
"source_idx": 4692,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7071,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7071
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004692",
"normalized_example_id": "ex_004692",
"source_idx": 4692,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7071,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 1159
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004695",
"normalized_example_id": "ex_004695",
"source_idx": 4695,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1932,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 1546
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004695",
"normalized_example_id": "ex_004695",
"source_idx": 4695,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1932,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 1932
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004695",
"normalized_example_id": "ex_004695",
"source_idx": 4695,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1932,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "135",
"budget": 3638
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004698",
"normalized_example_id": "ex_004698",
"source_idx": 4698,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6064,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "135",
"budget": 4851
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004698",
"normalized_example_id": "ex_004698",
"source_idx": 4698,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6064,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "135",
"budget": 6064
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004698",
"normalized_example_id": "ex_004698",
"source_idx": 4698,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6064,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{s_1}{9}",
"budget": 3956
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004701",
"normalized_example_id": "ex_004701",
"source_idx": 4701,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6594,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{s_1}{9}",
"budget": 5275
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004701",
"normalized_example_id": "ex_004701",
"source_idx": 4701,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6594,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{s_1}{9}",
"budget": 6594
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004701",
"normalized_example_id": "ex_004701",
"source_idx": 4701,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6594,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2151
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004702",
"normalized_example_id": "ex_004702",
"source_idx": 4702,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3585,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2868
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004702",
"normalized_example_id": "ex_004702",
"source_idx": 4702,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3585,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3585
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004702",
"normalized_example_id": "ex_004702",
"source_idx": 4702,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3585,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\arcsin\\frac{4}{5}",
"budget": 3317
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004704",
"normalized_example_id": "ex_004704",
"source_idx": 4704,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5529,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\arcsin\\frac{4}{5}",
"budget": 4423
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004704",
"normalized_example_id": "ex_004704",
"source_idx": 4704,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5529,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\arcsin\\frac{4}{5}",
"budget": 5529
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004704",
"normalized_example_id": "ex_004704",
"source_idx": 4704,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5529,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "156,37",
"budget": 4793
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004708",
"normalized_example_id": "ex_004708",
"source_idx": 4708,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7989,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "156,37",
"budget": 6391
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004708",
"normalized_example_id": "ex_004708",
"source_idx": 4708,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7989,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "156,37",
"budget": 7989
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004708",
"normalized_example_id": "ex_004708",
"source_idx": 4708,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7989,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "788",
"budget": 2996
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004710",
"normalized_example_id": "ex_004710",
"source_idx": 4710,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4994,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "788",
"budget": 3995
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004710",
"normalized_example_id": "ex_004710",
"source_idx": 4710,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4994,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "788",
"budget": 4994
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004710",
"normalized_example_id": "ex_004710",
"source_idx": 4710,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4994,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "D",
"budget": 1914
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004716",
"normalized_example_id": "ex_004716",
"source_idx": 4716,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3190,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "D",
"budget": 2552
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004716",
"normalized_example_id": "ex_004716",
"source_idx": 4716,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3190,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "D",
"budget": 3190
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004716",
"normalized_example_id": "ex_004716",
"source_idx": 4716,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3190,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(3,2,1)",
"budget": 5240
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004720",
"normalized_example_id": "ex_004720",
"source_idx": 4720,
"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": "(3,2,1)",
"budget": 6986
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004720",
"normalized_example_id": "ex_004720",
"source_idx": 4720,
"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": "(3,2,1)",
"budget": 8733
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004720",
"normalized_example_id": "ex_004720",
"source_idx": 4720,
"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": "m=1",
"budget": 4741
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004722",
"normalized_example_id": "ex_004722",
"source_idx": 4722,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7902,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "m=1",
"budget": 6322
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004722",
"normalized_example_id": "ex_004722",
"source_idx": 4722,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7902,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "m=1",
"budget": 7902
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004722",
"normalized_example_id": "ex_004722",
"source_idx": 4722,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7902,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x)=x+1",
"budget": 4419
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004724",
"normalized_example_id": "ex_004724",
"source_idx": 4724,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7365,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x)=x+1",
"budget": 5892
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004724",
"normalized_example_id": "ex_004724",
"source_idx": 4724,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7365,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x)=x+1",
"budget": 7365
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004724",
"normalized_example_id": "ex_004724",
"source_idx": 4724,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7365,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "580",
"budget": 2533
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004725",
"normalized_example_id": "ex_004725",
"source_idx": 4725,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4222,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "580",
"budget": 3378
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004725",
"normalized_example_id": "ex_004725",
"source_idx": 4725,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4222,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "580",
"budget": 4222
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004725",
"normalized_example_id": "ex_004725",
"source_idx": 4725,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4222,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{\\sqrt{6}}{2}",
"budget": 3364
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004727",
"normalized_example_id": "ex_004727",
"source_idx": 4727,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5606,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{\\sqrt{6}}{2}",
"budget": 4485
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004727",
"normalized_example_id": "ex_004727",
"source_idx": 4727,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5606,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{\\sqrt{6}}{2}",
"budget": 5606
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004727",
"normalized_example_id": "ex_004727",
"source_idx": 4727,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5606,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1208
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004729",
"normalized_example_id": "ex_004729",
"source_idx": 4729,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2013,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1610
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004729",
"normalized_example_id": "ex_004729",
"source_idx": 4729,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2013,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2013
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004729",
"normalized_example_id": "ex_004729",
"source_idx": 4729,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2013,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{5}",
"budget": 2037
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004731",
"normalized_example_id": "ex_004731",
"source_idx": 4731,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3395,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{5}",
"budget": 2716
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004731",
"normalized_example_id": "ex_004731",
"source_idx": 4731,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3395,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{5}",
"budget": 3395
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004731",
"normalized_example_id": "ex_004731",
"source_idx": 4731,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3395,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\alpha = -3",
"budget": 4219
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004733",
"normalized_example_id": "ex_004733",
"source_idx": 4733,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7031,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\alpha = -3",
"budget": 5625
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004733",
"normalized_example_id": "ex_004733",
"source_idx": 4733,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7031,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\alpha = -3",
"budget": 7031
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004733",
"normalized_example_id": "ex_004733",
"source_idx": 4733,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7031,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "100",
"budget": 1055
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004734",
"normalized_example_id": "ex_004734",
"source_idx": 4734,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1759,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.