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": "223",
"budget": 2913
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003074",
"normalized_example_id": "ex_003074",
"source_idx": 3074,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4855,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "223",
"budget": 3884
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003074",
"normalized_example_id": "ex_003074",
"source_idx": 3074,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4855,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "223",
"budget": 4855
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003074",
"normalized_example_id": "ex_003074",
"source_idx": 3074,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4855,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "53",
"budget": 3478
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003075",
"normalized_example_id": "ex_003075",
"source_idx": 3075,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5796,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "53",
"budget": 4637
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003075",
"normalized_example_id": "ex_003075",
"source_idx": 3075,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5796,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "53",
"budget": 5796
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003075",
"normalized_example_id": "ex_003075",
"source_idx": 3075,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5796,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1325
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003077",
"normalized_example_id": "ex_003077",
"source_idx": 3077,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003077",
"normalized_example_id": "ex_003077",
"source_idx": 3077,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2208
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003077",
"normalized_example_id": "ex_003077",
"source_idx": 3077,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "540",
"budget": 1733
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003078",
"normalized_example_id": "ex_003078",
"source_idx": 3078,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2888,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "540",
"budget": 2310
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003078",
"normalized_example_id": "ex_003078",
"source_idx": 3078,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2888,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "540",
"budget": 2888
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003078",
"normalized_example_id": "ex_003078",
"source_idx": 3078,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2888,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{25}{2}",
"budget": 1766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003080",
"normalized_example_id": "ex_003080",
"source_idx": 3080,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2943,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{25}{2}",
"budget": 2354
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003080",
"normalized_example_id": "ex_003080",
"source_idx": 3080,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2943,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{25}{2}",
"budget": 2943
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003080",
"normalized_example_id": "ex_003080",
"source_idx": 3080,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2943,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "71",
"budget": 3880
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003081",
"normalized_example_id": "ex_003081",
"source_idx": 3081,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6466,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "71",
"budget": 5173
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003081",
"normalized_example_id": "ex_003081",
"source_idx": 3081,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6466,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "71",
"budget": 6466
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003081",
"normalized_example_id": "ex_003081",
"source_idx": 3081,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6466,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 760
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003082",
"normalized_example_id": "ex_003082",
"source_idx": 3082,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1267,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1014
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003082",
"normalized_example_id": "ex_003082",
"source_idx": 3082,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1267,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1267
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003082",
"normalized_example_id": "ex_003082",
"source_idx": 3082,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1267,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "50",
"budget": 1147
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003083",
"normalized_example_id": "ex_003083",
"source_idx": 3083,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1911,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "50",
"budget": 1529
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003083",
"normalized_example_id": "ex_003083",
"source_idx": 3083,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1911,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "50",
"budget": 1911
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003083",
"normalized_example_id": "ex_003083",
"source_idx": 3083,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1911,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 818
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003085",
"normalized_example_id": "ex_003085",
"source_idx": 3085,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1363,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1090
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003085",
"normalized_example_id": "ex_003085",
"source_idx": 3085,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1363,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1363
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003085",
"normalized_example_id": "ex_003085",
"source_idx": 3085,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1363,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "314",
"budget": 1466
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003088",
"normalized_example_id": "ex_003088",
"source_idx": 3088,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2444,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "314",
"budget": 1955
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003088",
"normalized_example_id": "ex_003088",
"source_idx": 3088,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2444,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "314",
"budget": 2444
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003088",
"normalized_example_id": "ex_003088",
"source_idx": 3088,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2444,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{(2n-3)\\cdot3^{n}+3}{4}",
"budget": 2281
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003090",
"normalized_example_id": "ex_003090",
"source_idx": 3090,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3801,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{(2n-3)\\cdot3^{n}+3}{4}",
"budget": 3041
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003090",
"normalized_example_id": "ex_003090",
"source_idx": 3090,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3801,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{(2n-3)\\cdot3^{n}+3}{4}",
"budget": 3801
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003090",
"normalized_example_id": "ex_003090",
"source_idx": 3090,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3801,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "82",
"budget": 5021
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003097",
"normalized_example_id": "ex_003097",
"source_idx": 3097,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8369,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "82",
"budget": 6695
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003097",
"normalized_example_id": "ex_003097",
"source_idx": 3097,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8369,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "82",
"budget": 8369
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003097",
"normalized_example_id": "ex_003097",
"source_idx": 3097,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8369,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "433",
"budget": 3619
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003099",
"normalized_example_id": "ex_003099",
"source_idx": 3099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6032,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "433",
"budget": 4826
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003099",
"normalized_example_id": "ex_003099",
"source_idx": 3099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6032,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "433",
"budget": 6032
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003099",
"normalized_example_id": "ex_003099",
"source_idx": 3099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6032,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 2713
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003101",
"normalized_example_id": "ex_003101",
"source_idx": 3101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4521,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 3617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003101",
"normalized_example_id": "ex_003101",
"source_idx": 3101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4521,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 4521
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003101",
"normalized_example_id": "ex_003101",
"source_idx": 3101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4521,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "42",
"budget": 2275
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003103",
"normalized_example_id": "ex_003103",
"source_idx": 3103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3791,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "42",
"budget": 3033
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003103",
"normalized_example_id": "ex_003103",
"source_idx": 3103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3791,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "42",
"budget": 3791
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003103",
"normalized_example_id": "ex_003103",
"source_idx": 3103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3791,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(668,668,667),(668,667,668),(667,668,668)",
"budget": 3922
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003104",
"normalized_example_id": "ex_003104",
"source_idx": 3104,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6536,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(668,668,667),(668,667,668),(667,668,668)",
"budget": 5229
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003104",
"normalized_example_id": "ex_003104",
"source_idx": 3104,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6536,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(668,668,667),(668,667,668),(667,668,668)",
"budget": 6536
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003104",
"normalized_example_id": "ex_003104",
"source_idx": 3104,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6536,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2.002",
"budget": 2138
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003107",
"normalized_example_id": "ex_003107",
"source_idx": 3107,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3563,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2.002",
"budget": 2850
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003107",
"normalized_example_id": "ex_003107",
"source_idx": 3107,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3563,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2.002",
"budget": 3563
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003107",
"normalized_example_id": "ex_003107",
"source_idx": 3107,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3563,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,y)=(0,0)",
"budget": 5994
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003108",
"normalized_example_id": "ex_003108",
"source_idx": 3108,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9990,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,y)=(0,0)",
"budget": 7992
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003108",
"normalized_example_id": "ex_003108",
"source_idx": 3108,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9990,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,y)=(0,0)",
"budget": 9990
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003108",
"normalized_example_id": "ex_003108",
"source_idx": 3108,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9990,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2r",
"budget": 1766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003109",
"normalized_example_id": "ex_003109",
"source_idx": 3109,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2943,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2r",
"budget": 2354
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003109",
"normalized_example_id": "ex_003109",
"source_idx": 3109,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2943,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2r",
"budget": 2943
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003109",
"normalized_example_id": "ex_003109",
"source_idx": 3109,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2943,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1:1",
"budget": 2395
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003110",
"normalized_example_id": "ex_003110",
"source_idx": 3110,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3991,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1:1",
"budget": 3193
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003110",
"normalized_example_id": "ex_003110",
"source_idx": 3110,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3991,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1:1",
"budget": 3991
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003110",
"normalized_example_id": "ex_003110",
"source_idx": 3110,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3991,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1954
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003111",
"normalized_example_id": "ex_003111",
"source_idx": 3111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3256,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2605
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003111",
"normalized_example_id": "ex_003111",
"source_idx": 3111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3256,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3256
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003111",
"normalized_example_id": "ex_003111",
"source_idx": 3111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3256,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4646
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003113",
"normalized_example_id": "ex_003113",
"source_idx": 3113,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7743,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6194
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003113",
"normalized_example_id": "ex_003113",
"source_idx": 3113,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7743,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7743
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003113",
"normalized_example_id": "ex_003113",
"source_idx": 3113,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7743,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "F(x)=\\sqrt{\\ln^{2}x+1}",
"budget": 3188
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003114",
"normalized_example_id": "ex_003114",
"source_idx": 3114,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5313,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "F(x)=\\sqrt{\\ln^{2}x+1}",
"budget": 4250
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003114",
"normalized_example_id": "ex_003114",
"source_idx": 3114,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5313,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "F(x)=\\sqrt{\\ln^{2}x+1}",
"budget": 5313
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003114",
"normalized_example_id": "ex_003114",
"source_idx": 3114,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5313,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "p(x) = a(x^2 - 1)^2 + 1 - a",
"budget": 5617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003115",
"normalized_example_id": "ex_003115",
"source_idx": 3115,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9362,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "p(x) = a(x^2 - 1)^2 + 1 - a",
"budget": 7490
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003115",
"normalized_example_id": "ex_003115",
"source_idx": 3115,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9362,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "p(x) = a(x^2 - 1)^2 + 1 - a",
"budget": 9362
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003115",
"normalized_example_id": "ex_003115",
"source_idx": 3115,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9362,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2:3",
"budget": 2146
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003116",
"normalized_example_id": "ex_003116",
"source_idx": 3116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3576,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2:3",
"budget": 2861
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003116",
"normalized_example_id": "ex_003116",
"source_idx": 3116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3576,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2:3",
"budget": 3576
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003116",
"normalized_example_id": "ex_003116",
"source_idx": 3116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3576,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "630",
"budget": 3983
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003117",
"normalized_example_id": "ex_003117",
"source_idx": 3117,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6638,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "630",
"budget": 5310
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003117",
"normalized_example_id": "ex_003117",
"source_idx": 3117,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6638,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "630",
"budget": 6638
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003117",
"normalized_example_id": "ex_003117",
"source_idx": 3117,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6638,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 2589
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003118",
"normalized_example_id": "ex_003118",
"source_idx": 3118,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4315,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 3452
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003118",
"normalized_example_id": "ex_003118",
"source_idx": 3118,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4315,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6",
"budget": 4315
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003118",
"normalized_example_id": "ex_003118",
"source_idx": 3118,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4315,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "84",
"budget": 5465
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003119",
"normalized_example_id": "ex_003119",
"source_idx": 3119,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9109,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "84",
"budget": 7287
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003119",
"normalized_example_id": "ex_003119",
"source_idx": 3119,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9109,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "84",
"budget": 9109
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003119",
"normalized_example_id": "ex_003119",
"source_idx": 3119,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9109,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "220",
"budget": 1952
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003121",
"normalized_example_id": "ex_003121",
"source_idx": 3121,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3254,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "220",
"budget": 2603
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003121",
"normalized_example_id": "ex_003121",
"source_idx": 3121,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3254,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "220",
"budget": 3254
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003121",
"normalized_example_id": "ex_003121",
"source_idx": 3121,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3254,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8k+1",
"budget": 5909
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003123",
"normalized_example_id": "ex_003123",
"source_idx": 3123,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9848,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8k+1",
"budget": 7878
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003123",
"normalized_example_id": "ex_003123",
"source_idx": 3123,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9848,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8k+1",
"budget": 9848
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003123",
"normalized_example_id": "ex_003123",
"source_idx": 3123,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9848,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\alpha+\\frac{\\beta-\\alpha}{\\sqrt{e}}",
"budget": 4835
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003126",
"normalized_example_id": "ex_003126",
"source_idx": 3126,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8059,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\alpha+\\frac{\\beta-\\alpha}{\\sqrt{e}}",
"budget": 6447
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003126",
"normalized_example_id": "ex_003126",
"source_idx": 3126,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8059,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\alpha+\\frac{\\beta-\\alpha}{\\sqrt{e}}",
"budget": 8059
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003126",
"normalized_example_id": "ex_003126",
"source_idx": 3126,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8059,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(\\frac{a_{1}}{c_{1}}+\\frac{a_{2}}{c_{2}}+\\frac{a_{3}}{c_{3}})x+(\\frac{b_{1}}{c_{1}}+\\frac{b_{2}}{c_{2}}+\\frac{b_{3}}{c_{3}})y+1=0",
"budget": 2398
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003127",
"normalized_example_id": "ex_003127",
"source_idx": 3127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3997,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(\\frac{a_{1}}{c_{1}}+\\frac{a_{2}}{c_{2}}+\\frac{a_{3}}{c_{3}})x+(\\frac{b_{1}}{c_{1}}+\\frac{b_{2}}{c_{2}}+\\frac{b_{3}}{c_{3}})y+1=0",
"budget": 3198
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003127",
"normalized_example_id": "ex_003127",
"source_idx": 3127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3997,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(\\frac{a_{1}}{c_{1}}+\\frac{a_{2}}{c_{2}}+\\frac{a_{3}}{c_{3}})x+(\\frac{b_{1}}{c_{1}}+\\frac{b_{2}}{c_{2}}+\\frac{b_{3}}{c_{3}})y+1=0",
"budget": 3997
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003127",
"normalized_example_id": "ex_003127",
"source_idx": 3127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3997,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4017",
"budget": 3904
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003129",
"normalized_example_id": "ex_003129",
"source_idx": 3129,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6507,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4017",
"budget": 5206
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003129",
"normalized_example_id": "ex_003129",
"source_idx": 3129,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6507,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4017",
"budget": 6507
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003129",
"normalized_example_id": "ex_003129",
"source_idx": 3129,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6507,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "998910",
"budget": 3218
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003132",
"normalized_example_id": "ex_003132",
"source_idx": 3132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5363,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.