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": "2500",
"budget": 3536
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001938",
"normalized_example_id": "ex_001938",
"source_idx": 1938,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4420,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2500",
"budget": 4420
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001938",
"normalized_example_id": "ex_001938",
"source_idx": 1938,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4420,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5<\\nu<15",
"budget": 1623
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001939",
"normalized_example_id": "ex_001939",
"source_idx": 1939,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5<\\nu<15",
"budget": 2164
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001939",
"normalized_example_id": "ex_001939",
"source_idx": 1939,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5<\\nu<15",
"budget": 2705
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001939",
"normalized_example_id": "ex_001939",
"source_idx": 1939,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}\\lny+\\frac{1}{3}(y^{2}+1)^{3/2}=C",
"budget": 1400
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001941",
"normalized_example_id": "ex_001941",
"source_idx": 1941,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2334,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}\\lny+\\frac{1}{3}(y^{2}+1)^{3/2}=C",
"budget": 1867
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001941",
"normalized_example_id": "ex_001941",
"source_idx": 1941,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2334,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}\\lny+\\frac{1}{3}(y^{2}+1)^{3/2}=C",
"budget": 2334
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001941",
"normalized_example_id": "ex_001941",
"source_idx": 1941,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2334,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "20",
"budget": 2387
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001942",
"normalized_example_id": "ex_001942",
"source_idx": 1942,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3979,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "20",
"budget": 3183
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001942",
"normalized_example_id": "ex_001942",
"source_idx": 1942,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3979,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "20",
"budget": 3979
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001942",
"normalized_example_id": "ex_001942",
"source_idx": 1942,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3979,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1431
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001946",
"normalized_example_id": "ex_001946",
"source_idx": 1946,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2385,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1908
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001946",
"normalized_example_id": "ex_001946",
"source_idx": 1946,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2385,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2385
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001946",
"normalized_example_id": "ex_001946",
"source_idx": 1946,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2385,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2500",
"budget": 898
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001947",
"normalized_example_id": "ex_001947",
"source_idx": 1947,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1496,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2500",
"budget": 1197
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001947",
"normalized_example_id": "ex_001947",
"source_idx": 1947,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1496,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2500",
"budget": 1496
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001947",
"normalized_example_id": "ex_001947",
"source_idx": 1947,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1496,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "648",
"budget": 1042
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001948",
"normalized_example_id": "ex_001948",
"source_idx": 1948,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1736,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "648",
"budget": 1389
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001948",
"normalized_example_id": "ex_001948",
"source_idx": 1948,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1736,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "648",
"budget": 1736
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001948",
"normalized_example_id": "ex_001948",
"source_idx": 1948,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1736,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14,23,28",
"budget": 2619
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001953",
"normalized_example_id": "ex_001953",
"source_idx": 1953,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4365,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14,23,28",
"budget": 3492
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001953",
"normalized_example_id": "ex_001953",
"source_idx": 1953,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4365,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14,23,28",
"budget": 4365
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001953",
"normalized_example_id": "ex_001953",
"source_idx": 1953,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4365,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "52\\mathrm{~}",
"budget": 4828
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001955",
"normalized_example_id": "ex_001955",
"source_idx": 1955,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8046,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "52\\mathrm{~}",
"budget": 6437
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001955",
"normalized_example_id": "ex_001955",
"source_idx": 1955,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8046,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "52\\mathrm{~}",
"budget": 8046
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001955",
"normalized_example_id": "ex_001955",
"source_idx": 1955,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8046,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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=13,y=10",
"budget": 954
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001956",
"normalized_example_id": "ex_001956",
"source_idx": 1956,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1590,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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=13,y=10",
"budget": 1272
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001956",
"normalized_example_id": "ex_001956",
"source_idx": 1956,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1590,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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=13,y=10",
"budget": 1590
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001956",
"normalized_example_id": "ex_001956",
"source_idx": 1956,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1590,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "94",
"budget": 1033
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001957",
"normalized_example_id": "ex_001957",
"source_idx": 1957,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1721,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "94",
"budget": 1377
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001957",
"normalized_example_id": "ex_001957",
"source_idx": 1957,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1721,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "94",
"budget": 1721
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001957",
"normalized_example_id": "ex_001957",
"source_idx": 1957,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1721,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1\\frac{3}{4}",
"budget": 2692
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001960",
"normalized_example_id": "ex_001960",
"source_idx": 1960,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4486,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1\\frac{3}{4}",
"budget": 3589
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001960",
"normalized_example_id": "ex_001960",
"source_idx": 1960,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4486,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1\\frac{3}{4}",
"budget": 4486
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001960",
"normalized_example_id": "ex_001960",
"source_idx": 1960,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4486,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "490",
"budget": 3073
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001961",
"normalized_example_id": "ex_001961",
"source_idx": 1961,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5121,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "490",
"budget": 4097
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001961",
"normalized_example_id": "ex_001961",
"source_idx": 1961,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5121,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "490",
"budget": 5121
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001961",
"normalized_example_id": "ex_001961",
"source_idx": 1961,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5121,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2400",
"budget": 1700
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001962",
"normalized_example_id": "ex_001962",
"source_idx": 1962,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2833,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2400",
"budget": 2266
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001962",
"normalized_example_id": "ex_001962",
"source_idx": 1962,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2833,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2400",
"budget": 2833
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001962",
"normalized_example_id": "ex_001962",
"source_idx": 1962,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2833,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,2),(6,2)",
"budget": 5182
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001963",
"normalized_example_id": "ex_001963",
"source_idx": 1963,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8637,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,2),(6,2)",
"budget": 6910
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001963",
"normalized_example_id": "ex_001963",
"source_idx": 1963,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8637,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,2),(6,2)",
"budget": 8637
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001963",
"normalized_example_id": "ex_001963",
"source_idx": 1963,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8637,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2024 \\frac{3}{4}",
"budget": 4660
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001965",
"normalized_example_id": "ex_001965",
"source_idx": 1965,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2024 \\frac{3}{4}",
"budget": 6213
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001965",
"normalized_example_id": "ex_001965",
"source_idx": 1965,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2024 \\frac{3}{4}",
"budget": 7766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001965",
"normalized_example_id": "ex_001965",
"source_idx": 1965,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[\\frac{1}{2}(\\sqrt{8n+1}-1)]",
"budget": 2510
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001968",
"normalized_example_id": "ex_001968",
"source_idx": 1968,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4184,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[\\frac{1}{2}(\\sqrt{8n+1}-1)]",
"budget": 3347
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001968",
"normalized_example_id": "ex_001968",
"source_idx": 1968,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4184,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[\\frac{1}{2}(\\sqrt{8n+1}-1)]",
"budget": 4184
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001968",
"normalized_example_id": "ex_001968",
"source_idx": 1968,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4184,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001969",
"normalized_example_id": "ex_001969",
"source_idx": 1969,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1276,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1021
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001969",
"normalized_example_id": "ex_001969",
"source_idx": 1969,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1276,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1276
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001969",
"normalized_example_id": "ex_001969",
"source_idx": 1969,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1276,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;4)",
"budget": 4071
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001971",
"normalized_example_id": "ex_001971",
"source_idx": 1971,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6785,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;4)",
"budget": 5428
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001971",
"normalized_example_id": "ex_001971",
"source_idx": 1971,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6785,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;4)",
"budget": 6785
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001971",
"normalized_example_id": "ex_001971",
"source_idx": 1971,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6785,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "106",
"budget": 3278
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001973",
"normalized_example_id": "ex_001973",
"source_idx": 1973,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5464,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "106",
"budget": 4371
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001973",
"normalized_example_id": "ex_001973",
"source_idx": 1973,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5464,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "106",
"budget": 5464
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001973",
"normalized_example_id": "ex_001973",
"source_idx": 1973,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5464,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "155",
"budget": 1885
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001975",
"normalized_example_id": "ex_001975",
"source_idx": 1975,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3142,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "155",
"budget": 2514
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001975",
"normalized_example_id": "ex_001975",
"source_idx": 1975,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3142,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "155",
"budget": 3142
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001975",
"normalized_example_id": "ex_001975",
"source_idx": 1975,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3142,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "66",
"budget": 1055
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001977",
"normalized_example_id": "ex_001977",
"source_idx": 1977,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1758,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "66",
"budget": 1406
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001977",
"normalized_example_id": "ex_001977",
"source_idx": 1977,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1758,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "66",
"budget": 1758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001977",
"normalized_example_id": "ex_001977",
"source_idx": 1977,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1758,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "80,98",
"budget": 2315
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001978",
"normalized_example_id": "ex_001978",
"source_idx": 1978,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3858,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "80,98",
"budget": 3086
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001978",
"normalized_example_id": "ex_001978",
"source_idx": 1978,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3858,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "80,98",
"budget": 3858
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001978",
"normalized_example_id": "ex_001978",
"source_idx": 1978,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3858,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{e^{x}}{e^{2x}-6e^{x}+13}",
"budget": 1547
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001979",
"normalized_example_id": "ex_001979",
"source_idx": 1979,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2579,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{e^{x}}{e^{2x}-6e^{x}+13}",
"budget": 2063
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001979",
"normalized_example_id": "ex_001979",
"source_idx": 1979,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2579,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{e^{x}}{e^{2x}-6e^{x}+13}",
"budget": 2579
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001979",
"normalized_example_id": "ex_001979",
"source_idx": 1979,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2579,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 3658
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001982",
"normalized_example_id": "ex_001982",
"source_idx": 1982,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6097,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 4878
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001982",
"normalized_example_id": "ex_001982",
"source_idx": 1982,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6097,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 6097
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001982",
"normalized_example_id": "ex_001982",
"source_idx": 1982,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6097,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "570",
"budget": 2686
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001985",
"normalized_example_id": "ex_001985",
"source_idx": 1985,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4476,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "570",
"budget": 3581
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001985",
"normalized_example_id": "ex_001985",
"source_idx": 1985,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4476,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "570",
"budget": 4476
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001985",
"normalized_example_id": "ex_001985",
"source_idx": 1985,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4476,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1200",
"budget": 3254
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001988",
"normalized_example_id": "ex_001988",
"source_idx": 1988,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5424,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1200",
"budget": 4339
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001988",
"normalized_example_id": "ex_001988",
"source_idx": 1988,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5424,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1200",
"budget": 5424
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001988",
"normalized_example_id": "ex_001988",
"source_idx": 1988,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5424,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3775",
"budget": 893
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001990",
"normalized_example_id": "ex_001990",
"source_idx": 1990,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1489,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3775",
"budget": 1191
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001990",
"normalized_example_id": "ex_001990",
"source_idx": 1990,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1489,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3775",
"budget": 1489
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001990",
"normalized_example_id": "ex_001990",
"source_idx": 1990,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1489,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9:10",
"budget": 5067
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001991",
"normalized_example_id": "ex_001991",
"source_idx": 1991,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8445,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9:10",
"budget": 6756
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001991",
"normalized_example_id": "ex_001991",
"source_idx": 1991,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8445,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9:10",
"budget": 8445
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001991",
"normalized_example_id": "ex_001991",
"source_idx": 1991,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8445,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 (-1)^n n! + 1",
"budget": 3918
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001994",
"normalized_example_id": "ex_001994",
"source_idx": 1994,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6530,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 (-1)^n n! + 1",
"budget": 5224
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001994",
"normalized_example_id": "ex_001994",
"source_idx": 1994,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6530,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 (-1)^n n! + 1",
"budget": 6530
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001994",
"normalized_example_id": "ex_001994",
"source_idx": 1994,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6530,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3016
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001995",
"normalized_example_id": "ex_001995",
"source_idx": 1995,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5026,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4021
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001995",
"normalized_example_id": "ex_001995",
"source_idx": 1995,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5026,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5026
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001995",
"normalized_example_id": "ex_001995",
"source_idx": 1995,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5026,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1151
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001997",
"normalized_example_id": "ex_001997",
"source_idx": 1997,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1918,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1534
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001997",
"normalized_example_id": "ex_001997",
"source_idx": 1997,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1918,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1918
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001997",
"normalized_example_id": "ex_001997",
"source_idx": 1997,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1918,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "27",
"budget": 1670
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001998",
"normalized_example_id": "ex_001998",
"source_idx": 1998,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2783,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "27",
"budget": 2226
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001998",
"normalized_example_id": "ex_001998",
"source_idx": 1998,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2783,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "27",
"budget": 2783
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_001998",
"normalized_example_id": "ex_001998",
"source_idx": 1998,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2783,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\text{B}",
"budget": 3524
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002001",
"normalized_example_id": "ex_002001",
"source_idx": 2001,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5873,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\text{B}",
"budget": 4698
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002001",
"normalized_example_id": "ex_002001",
"source_idx": 2001,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5873,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.