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": "48",
"budget": 2919
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002928",
"normalized_example_id": "ex_002928",
"source_idx": 2928,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3649,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "48",
"budget": 3649
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002928",
"normalized_example_id": "ex_002928",
"source_idx": 2928,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3649,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8,12,18,27",
"budget": 799
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002929",
"normalized_example_id": "ex_002929",
"source_idx": 2929,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1331,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8,12,18,27",
"budget": 1065
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002929",
"normalized_example_id": "ex_002929",
"source_idx": 2929,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1331,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8,12,18,27",
"budget": 1331
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002929",
"normalized_example_id": "ex_002929",
"source_idx": 2929,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1331,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3k^2",
"budget": 5996
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002933",
"normalized_example_id": "ex_002933",
"source_idx": 2933,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9994,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3k^2",
"budget": 7995
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002933",
"normalized_example_id": "ex_002933",
"source_idx": 2933,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9994,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3k^2",
"budget": 9994
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002933",
"normalized_example_id": "ex_002933",
"source_idx": 2933,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9994,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2",
"budget": 5823
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002936",
"normalized_example_id": "ex_002936",
"source_idx": 2936,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2",
"budget": 7764
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002936",
"normalized_example_id": "ex_002936",
"source_idx": 2936,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2",
"budget": 9705
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002936",
"normalized_example_id": "ex_002936",
"source_idx": 2936,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, 4), (4, 2, 1), (3, 3, 3)",
"budget": 4609
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002940",
"normalized_example_id": "ex_002940",
"source_idx": 2940,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7682,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, 4), (4, 2, 1), (3, 3, 3)",
"budget": 6146
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002940",
"normalized_example_id": "ex_002940",
"source_idx": 2940,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7682,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, 4), (4, 2, 1), (3, 3, 3)",
"budget": 7682
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002940",
"normalized_example_id": "ex_002940",
"source_idx": 2940,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7682,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}+y^{2}-\\sqrt{3} xy=\\frac{a^{2}}{4}",
"budget": 4006
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002941",
"normalized_example_id": "ex_002941",
"source_idx": 2941,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6676,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}+y^{2}-\\sqrt{3} xy=\\frac{a^{2}}{4}",
"budget": 5341
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002941",
"normalized_example_id": "ex_002941",
"source_idx": 2941,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6676,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}+y^{2}-\\sqrt{3} xy=\\frac{a^{2}}{4}",
"budget": 6676
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002941",
"normalized_example_id": "ex_002941",
"source_idx": 2941,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6676,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1304",
"budget": 5026
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002944",
"normalized_example_id": "ex_002944",
"source_idx": 2944,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1304",
"budget": 6702
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002944",
"normalized_example_id": "ex_002944",
"source_idx": 2944,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1304",
"budget": 8377
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002944",
"normalized_example_id": "ex_002944",
"source_idx": 2944,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8377,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "68",
"budget": 863
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002946",
"normalized_example_id": "ex_002946",
"source_idx": 2946,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1439,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "68",
"budget": 1151
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002946",
"normalized_example_id": "ex_002946",
"source_idx": 2946,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1439,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "68",
"budget": 1439
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002946",
"normalized_example_id": "ex_002946",
"source_idx": 2946,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1439,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4",
"budget": 2959
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002947",
"normalized_example_id": "ex_002947",
"source_idx": 2947,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4931,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4",
"budget": 3945
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002947",
"normalized_example_id": "ex_002947",
"source_idx": 2947,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4931,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4",
"budget": 4931
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002947",
"normalized_example_id": "ex_002947",
"source_idx": 2947,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4931,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 970
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002950",
"normalized_example_id": "ex_002950",
"source_idx": 2950,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1617,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 1294
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002950",
"normalized_example_id": "ex_002950",
"source_idx": 2950,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1617,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 1617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002950",
"normalized_example_id": "ex_002950",
"source_idx": 2950,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1617,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "972",
"budget": 5398
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002951",
"normalized_example_id": "ex_002951",
"source_idx": 2951,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8996,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "972",
"budget": 7197
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002951",
"normalized_example_id": "ex_002951",
"source_idx": 2951,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8996,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "972",
"budget": 8996
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002951",
"normalized_example_id": "ex_002951",
"source_idx": 2951,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8996,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "253",
"budget": 2272
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002959",
"normalized_example_id": "ex_002959",
"source_idx": 2959,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3786,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "253",
"budget": 3029
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002959",
"normalized_example_id": "ex_002959",
"source_idx": 2959,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3786,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "253",
"budget": 3786
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002959",
"normalized_example_id": "ex_002959",
"source_idx": 2959,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3786,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 842
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002960",
"normalized_example_id": "ex_002960",
"source_idx": 2960,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1404,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 1123
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002960",
"normalized_example_id": "ex_002960",
"source_idx": 2960,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1404,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 1404
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002960",
"normalized_example_id": "ex_002960",
"source_idx": 2960,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1404,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 4057
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002964",
"normalized_example_id": "ex_002964",
"source_idx": 2964,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6762,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 5410
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002964",
"normalized_example_id": "ex_002964",
"source_idx": 2964,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6762,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 6762
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002964",
"normalized_example_id": "ex_002964",
"source_idx": 2964,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6762,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,G,H",
"budget": 4817
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002966",
"normalized_example_id": "ex_002966",
"source_idx": 2966,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8029,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,G,H",
"budget": 6423
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002966",
"normalized_example_id": "ex_002966",
"source_idx": 2966,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8029,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,G,H",
"budget": 8029
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002966",
"normalized_example_id": "ex_002966",
"source_idx": 2966,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8029,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2543
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002967",
"normalized_example_id": "ex_002967",
"source_idx": 2967,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4239,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3391
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002967",
"normalized_example_id": "ex_002967",
"source_idx": 2967,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4239,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4239
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002967",
"normalized_example_id": "ex_002967",
"source_idx": 2967,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4239,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "28035",
"budget": 2917
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002971",
"normalized_example_id": "ex_002971",
"source_idx": 2971,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4862,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "28035",
"budget": 3890
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002971",
"normalized_example_id": "ex_002971",
"source_idx": 2971,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4862,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "28035",
"budget": 4862
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002971",
"normalized_example_id": "ex_002971",
"source_idx": 2971,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4862,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "C",
"budget": 5199
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002973",
"normalized_example_id": "ex_002973",
"source_idx": 2973,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8665,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "C",
"budget": 6932
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002973",
"normalized_example_id": "ex_002973",
"source_idx": 2973,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8665,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "C",
"budget": 8665
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002973",
"normalized_example_id": "ex_002973",
"source_idx": 2973,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8665,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 3933
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002974",
"normalized_example_id": "ex_002974",
"source_idx": 2974,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6555,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 5244
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002974",
"normalized_example_id": "ex_002974",
"source_idx": 2974,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6555,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 6555
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002974",
"normalized_example_id": "ex_002974",
"source_idx": 2974,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6555,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "80",
"budget": 1818
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002975",
"normalized_example_id": "ex_002975",
"source_idx": 2975,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3030,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "80",
"budget": 2424
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002975",
"normalized_example_id": "ex_002975",
"source_idx": 2975,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3030,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "80",
"budget": 3030
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002975",
"normalized_example_id": "ex_002975",
"source_idx": 2975,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3030,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3n^{2}+3n+1",
"budget": 2621
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002976",
"normalized_example_id": "ex_002976",
"source_idx": 2976,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4368,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3n^{2}+3n+1",
"budget": 3494
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002976",
"normalized_example_id": "ex_002976",
"source_idx": 2976,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4368,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3n^{2}+3n+1",
"budget": 4368
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002976",
"normalized_example_id": "ex_002976",
"source_idx": 2976,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4368,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "s_{n}=s_{0}r^{n}",
"budget": 1021
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002977",
"normalized_example_id": "ex_002977",
"source_idx": 2977,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1701,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "s_{n}=s_{0}r^{n}",
"budget": 1361
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002977",
"normalized_example_id": "ex_002977",
"source_idx": 2977,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1701,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "s_{n}=s_{0}r^{n}",
"budget": 1701
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002977",
"normalized_example_id": "ex_002977",
"source_idx": 2977,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1701,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3805
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002978",
"normalized_example_id": "ex_002978",
"source_idx": 2978,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6341,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5073
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002978",
"normalized_example_id": "ex_002978",
"source_idx": 2978,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6341,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6341
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002978",
"normalized_example_id": "ex_002978",
"source_idx": 2978,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6341,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6,7,8,9",
"budget": 1984
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002979",
"normalized_example_id": "ex_002979",
"source_idx": 2979,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3307,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6,7,8,9",
"budget": 2646
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002979",
"normalized_example_id": "ex_002979",
"source_idx": 2979,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3307,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6,7,8,9",
"budget": 3307
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002979",
"normalized_example_id": "ex_002979",
"source_idx": 2979,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3307,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1}{3},-\\frac{3}{8},-\\frac{2}{5}",
"budget": 2824
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002980",
"normalized_example_id": "ex_002980",
"source_idx": 2980,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4707,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1}{3},-\\frac{3}{8},-\\frac{2}{5}",
"budget": 3766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002980",
"normalized_example_id": "ex_002980",
"source_idx": 2980,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4707,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1}{3},-\\frac{3}{8},-\\frac{2}{5}",
"budget": 4707
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002980",
"normalized_example_id": "ex_002980",
"source_idx": 2980,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4707,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 4352
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002981",
"normalized_example_id": "ex_002981",
"source_idx": 2981,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7253,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 5802
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002981",
"normalized_example_id": "ex_002981",
"source_idx": 2981,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7253,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 7253
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002981",
"normalized_example_id": "ex_002981",
"source_idx": 2981,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7253,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "247",
"budget": 3280
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002983",
"normalized_example_id": "ex_002983",
"source_idx": 2983,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "247",
"budget": 4374
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002983",
"normalized_example_id": "ex_002983",
"source_idx": 2983,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "247",
"budget": 5467
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002983",
"normalized_example_id": "ex_002983",
"source_idx": 2983,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = \\frac{(-1)^{n-1}}{2^n}",
"budget": 5284
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002984",
"normalized_example_id": "ex_002984",
"source_idx": 2984,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8807,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = \\frac{(-1)^{n-1}}{2^n}",
"budget": 7046
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002984",
"normalized_example_id": "ex_002984",
"source_idx": 2984,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8807,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = \\frac{(-1)^{n-1}}{2^n}",
"budget": 8807
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002984",
"normalized_example_id": "ex_002984",
"source_idx": 2984,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8807,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-4,4]",
"budget": 2977
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002985",
"normalized_example_id": "ex_002985",
"source_idx": 2985,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4962,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-4,4]",
"budget": 3970
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002985",
"normalized_example_id": "ex_002985",
"source_idx": 2985,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4962,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-4,4]",
"budget": 4962
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002985",
"normalized_example_id": "ex_002985",
"source_idx": 2985,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4962,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4:21",
"budget": 4266
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002988",
"normalized_example_id": "ex_002988",
"source_idx": 2988,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7110,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4:21",
"budget": 5688
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002988",
"normalized_example_id": "ex_002988",
"source_idx": 2988,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7110,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4:21",
"budget": 7110
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002988",
"normalized_example_id": "ex_002988",
"source_idx": 2988,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7110,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3645",
"budget": 2584
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002989",
"normalized_example_id": "ex_002989",
"source_idx": 2989,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4307,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3645",
"budget": 3446
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002989",
"normalized_example_id": "ex_002989",
"source_idx": 2989,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4307,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3645",
"budget": 4307
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002989",
"normalized_example_id": "ex_002989",
"source_idx": 2989,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4307,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{5}{2}",
"budget": 6097
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002991",
"normalized_example_id": "ex_002991",
"source_idx": 2991,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10162,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{5}{2}",
"budget": 8130
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002991",
"normalized_example_id": "ex_002991",
"source_idx": 2991,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10162,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{5}{2}",
"budget": 10162
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002991",
"normalized_example_id": "ex_002991",
"source_idx": 2991,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10162,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[1,2,8],[1,2,15],[1,3,10],[1,5,6],[2,3,5]",
"budget": 4678
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002994",
"normalized_example_id": "ex_002994",
"source_idx": 2994,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7796,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[1,2,8],[1,2,15],[1,3,10],[1,5,6],[2,3,5]",
"budget": 6237
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002994",
"normalized_example_id": "ex_002994",
"source_idx": 2994,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7796,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[1,2,8],[1,2,15],[1,3,10],[1,5,6],[2,3,5]",
"budget": 7796
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002994",
"normalized_example_id": "ex_002994",
"source_idx": 2994,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7796,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1895",
"budget": 3972
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002996",
"normalized_example_id": "ex_002996",
"source_idx": 2996,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6620,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1895",
"budget": 5296
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002996",
"normalized_example_id": "ex_002996",
"source_idx": 2996,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6620,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.