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": "11",
"budget": 2019
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002304",
"normalized_example_id": "ex_002304",
"source_idx": 2304,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2524,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "11",
"budget": 2524
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002304",
"normalized_example_id": "ex_002304",
"source_idx": 2304,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2524,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1,2",
"budget": 4695
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002305",
"normalized_example_id": "ex_002305",
"source_idx": 2305,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7825,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1,2",
"budget": 6260
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002305",
"normalized_example_id": "ex_002305",
"source_idx": 2305,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7825,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1,2",
"budget": 7825
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002305",
"normalized_example_id": "ex_002305",
"source_idx": 2305,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7825,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2080
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002308",
"normalized_example_id": "ex_002308",
"source_idx": 2308,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2774
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002308",
"normalized_example_id": "ex_002308",
"source_idx": 2308,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3467
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002308",
"normalized_example_id": "ex_002308",
"source_idx": 2308,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3467,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "164",
"budget": 1345
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002313",
"normalized_example_id": "ex_002313",
"source_idx": 2313,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2241,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "164",
"budget": 1793
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002313",
"normalized_example_id": "ex_002313",
"source_idx": 2313,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2241,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "164",
"budget": 2241
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002313",
"normalized_example_id": "ex_002313",
"source_idx": 2313,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2241,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{3}{8}",
"budget": 1506
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002314",
"normalized_example_id": "ex_002314",
"source_idx": 2314,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2510,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{3}{8}",
"budget": 2008
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002314",
"normalized_example_id": "ex_002314",
"source_idx": 2314,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2510,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{3}{8}",
"budget": 2510
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002314",
"normalized_example_id": "ex_002314",
"source_idx": 2314,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2510,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-\\frac{5}{3},1)",
"budget": 631
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002316",
"normalized_example_id": "ex_002316",
"source_idx": 2316,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1051,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-\\frac{5}{3},1)",
"budget": 841
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002316",
"normalized_example_id": "ex_002316",
"source_idx": 2316,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1051,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-\\frac{5}{3},1)",
"budget": 1051
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002316",
"normalized_example_id": "ex_002316",
"source_idx": 2316,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1051,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4925
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002318",
"normalized_example_id": "ex_002318",
"source_idx": 2318,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6566
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002318",
"normalized_example_id": "ex_002318",
"source_idx": 2318,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 8208
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002318",
"normalized_example_id": "ex_002318",
"source_idx": 2318,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2223
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002319",
"normalized_example_id": "ex_002319",
"source_idx": 2319,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2964
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002319",
"normalized_example_id": "ex_002319",
"source_idx": 2319,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3705
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002319",
"normalized_example_id": "ex_002319",
"source_idx": 2319,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3705,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3279
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002321",
"normalized_example_id": "ex_002321",
"source_idx": 2321,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5465,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4372
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002321",
"normalized_example_id": "ex_002321",
"source_idx": 2321,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5465,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5465
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002321",
"normalized_example_id": "ex_002321",
"source_idx": 2321,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5465,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4801
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002323",
"normalized_example_id": "ex_002323",
"source_idx": 2323,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8002,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6402
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002323",
"normalized_example_id": "ex_002323",
"source_idx": 2323,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8002,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 8002
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002323",
"normalized_example_id": "ex_002323",
"source_idx": 2323,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8002,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{8}{\\sqrt{3}},\\frac{26}{\\sqrt{3}},\\frac{30}{\\sqrt{3}}",
"budget": 1591
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002324",
"normalized_example_id": "ex_002324",
"source_idx": 2324,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2651,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{8}{\\sqrt{3}},\\frac{26}{\\sqrt{3}},\\frac{30}{\\sqrt{3}}",
"budget": 2121
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002324",
"normalized_example_id": "ex_002324",
"source_idx": 2324,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2651,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{8}{\\sqrt{3}},\\frac{26}{\\sqrt{3}},\\frac{30}{\\sqrt{3}}",
"budget": 2651
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002324",
"normalized_example_id": "ex_002324",
"source_idx": 2324,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2651,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "672",
"budget": 1163
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002326",
"normalized_example_id": "ex_002326",
"source_idx": 2326,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1938,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "672",
"budget": 1550
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002326",
"normalized_example_id": "ex_002326",
"source_idx": 2326,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1938,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "672",
"budget": 1938
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002326",
"normalized_example_id": "ex_002326",
"source_idx": 2326,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1938,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7000",
"budget": 5538
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002327",
"normalized_example_id": "ex_002327",
"source_idx": 2327,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9230,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7000",
"budget": 7384
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002327",
"normalized_example_id": "ex_002327",
"source_idx": 2327,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9230,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7000",
"budget": 9230
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002327",
"normalized_example_id": "ex_002327",
"source_idx": 2327,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9230,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2\\sqrt{19};\\sqrt{37}",
"budget": 2525
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002328",
"normalized_example_id": "ex_002328",
"source_idx": 2328,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2\\sqrt{19};\\sqrt{37}",
"budget": 3366
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002328",
"normalized_example_id": "ex_002328",
"source_idx": 2328,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2\\sqrt{19};\\sqrt{37}",
"budget": 4208
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002328",
"normalized_example_id": "ex_002328",
"source_idx": 2328,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4208,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "127.40\\,",
"budget": 3392
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002330",
"normalized_example_id": "ex_002330",
"source_idx": 2330,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5654,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "127.40\\,",
"budget": 4523
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002330",
"normalized_example_id": "ex_002330",
"source_idx": 2330,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5654,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "127.40\\,",
"budget": 5654
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002330",
"normalized_example_id": "ex_002330",
"source_idx": 2330,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5654,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "325",
"budget": 4631
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002333",
"normalized_example_id": "ex_002333",
"source_idx": 2333,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7718,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "325",
"budget": 6174
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002333",
"normalized_example_id": "ex_002333",
"source_idx": 2333,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7718,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "325",
"budget": 7718
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002333",
"normalized_example_id": "ex_002333",
"source_idx": 2333,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7718,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 2882
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002334",
"normalized_example_id": "ex_002334",
"source_idx": 2334,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4803,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 3842
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002334",
"normalized_example_id": "ex_002334",
"source_idx": 2334,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4803,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 4803
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002334",
"normalized_example_id": "ex_002334",
"source_idx": 2334,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4803,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 920
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002335",
"normalized_example_id": "ex_002335",
"source_idx": 2335,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1533,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1226
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002335",
"normalized_example_id": "ex_002335",
"source_idx": 2335,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1533,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 1533
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002335",
"normalized_example_id": "ex_002335",
"source_idx": 2335,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1533,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2137
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002340",
"normalized_example_id": "ex_002340",
"source_idx": 2340,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3562,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2850
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002340",
"normalized_example_id": "ex_002340",
"source_idx": 2340,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3562,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3562
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002340",
"normalized_example_id": "ex_002340",
"source_idx": 2340,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3562,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, r) = (3, 2)",
"budget": 3978
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002341",
"normalized_example_id": "ex_002341",
"source_idx": 2341,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6630,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, r) = (3, 2)",
"budget": 5304
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002341",
"normalized_example_id": "ex_002341",
"source_idx": 2341,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6630,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, r) = (3, 2)",
"budget": 6630
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002341",
"normalized_example_id": "ex_002341",
"source_idx": 2341,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6630,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+\\sqrt{2}",
"budget": 4129
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002342",
"normalized_example_id": "ex_002342",
"source_idx": 2342,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6882,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+\\sqrt{2}",
"budget": 5506
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002342",
"normalized_example_id": "ex_002342",
"source_idx": 2342,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6882,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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+\\sqrt{2}",
"budget": 6882
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002342",
"normalized_example_id": "ex_002342",
"source_idx": 2342,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6882,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5629
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002343",
"normalized_example_id": "ex_002343",
"source_idx": 2343,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9382,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7506
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002343",
"normalized_example_id": "ex_002343",
"source_idx": 2343,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9382,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 9382
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002343",
"normalized_example_id": "ex_002343",
"source_idx": 2343,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9382,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "671",
"budget": 1463
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002344",
"normalized_example_id": "ex_002344",
"source_idx": 2344,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2438,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "671",
"budget": 1950
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002344",
"normalized_example_id": "ex_002344",
"source_idx": 2344,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2438,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "671",
"budget": 2438
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002344",
"normalized_example_id": "ex_002344",
"source_idx": 2344,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2438,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2035153",
"budget": 2423
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002348",
"normalized_example_id": "ex_002348",
"source_idx": 2348,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4039,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2035153",
"budget": 3231
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002348",
"normalized_example_id": "ex_002348",
"source_idx": 2348,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4039,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2035153",
"budget": 4039
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002348",
"normalized_example_id": "ex_002348",
"source_idx": 2348,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4039,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 790
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002352",
"normalized_example_id": "ex_002352",
"source_idx": 2352,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1316,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 1053
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002352",
"normalized_example_id": "ex_002352",
"source_idx": 2352,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1316,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 1316
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002352",
"normalized_example_id": "ex_002352",
"source_idx": 2352,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1316,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a > n^4 - n^3",
"budget": 5511
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002354",
"normalized_example_id": "ex_002354",
"source_idx": 2354,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9185,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a > n^4 - n^3",
"budget": 7348
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002354",
"normalized_example_id": "ex_002354",
"source_idx": 2354,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9185,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a > n^4 - n^3",
"budget": 9185
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002354",
"normalized_example_id": "ex_002354",
"source_idx": 2354,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9185,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(0,0,0,0) \\text{ and } \\left(\\frac{1}{3}, \\frac{1}{3}, \\frac{1}{3}, \\frac{1}{3}\\right)",
"budget": 4119
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002355",
"normalized_example_id": "ex_002355",
"source_idx": 2355,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6865,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(0,0,0,0) \\text{ and } \\left(\\frac{1}{3}, \\frac{1}{3}, \\frac{1}{3}, \\frac{1}{3}\\right)",
"budget": 5492
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002355",
"normalized_example_id": "ex_002355",
"source_idx": 2355,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6865,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(0,0,0,0) \\text{ and } \\left(\\frac{1}{3}, \\frac{1}{3}, \\frac{1}{3}, \\frac{1}{3}\\right)",
"budget": 6865
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002355",
"normalized_example_id": "ex_002355",
"source_idx": 2355,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6865,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n_{1}=\\cdots=n_{2017}=29",
"budget": 1052
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002356",
"normalized_example_id": "ex_002356",
"source_idx": 2356,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1753,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n_{1}=\\cdots=n_{2017}=29",
"budget": 1402
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002356",
"normalized_example_id": "ex_002356",
"source_idx": 2356,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1753,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n_{1}=\\cdots=n_{2017}=29",
"budget": 1753
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002356",
"normalized_example_id": "ex_002356",
"source_idx": 2356,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1753,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{\\sqrt{2}}-\\frac{1}{\\sqrt{2014}}",
"budget": 5260
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002357",
"normalized_example_id": "ex_002357",
"source_idx": 2357,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{\\sqrt{2}}-\\frac{1}{\\sqrt{2014}}",
"budget": 7013
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002357",
"normalized_example_id": "ex_002357",
"source_idx": 2357,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{\\sqrt{2}}-\\frac{1}{\\sqrt{2014}}",
"budget": 8766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002357",
"normalized_example_id": "ex_002357",
"source_idx": 2357,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,z)=(8,7,13)",
"budget": 5677
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002358",
"normalized_example_id": "ex_002358",
"source_idx": 2358,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9462,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,z)=(8,7,13)",
"budget": 7570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002358",
"normalized_example_id": "ex_002358",
"source_idx": 2358,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9462,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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,z)=(8,7,13)",
"budget": 9462
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002358",
"normalized_example_id": "ex_002358",
"source_idx": 2358,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9462,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7.5",
"budget": 1177
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002360",
"normalized_example_id": "ex_002360",
"source_idx": 2360,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1962,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7.5",
"budget": 1570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002360",
"normalized_example_id": "ex_002360",
"source_idx": 2360,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1962,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7.5",
"budget": 1962
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002360",
"normalized_example_id": "ex_002360",
"source_idx": 2360,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1962,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{4}{5}",
"budget": 2022
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002363",
"normalized_example_id": "ex_002363",
"source_idx": 2363,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3370,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{4}{5}",
"budget": 2696
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002363",
"normalized_example_id": "ex_002363",
"source_idx": 2363,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3370,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{4}{5}",
"budget": 3370
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002363",
"normalized_example_id": "ex_002363",
"source_idx": 2363,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3370,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5251
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002367",
"normalized_example_id": "ex_002367",
"source_idx": 2367,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8751,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7001
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002367",
"normalized_example_id": "ex_002367",
"source_idx": 2367,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8751,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 8751
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002367",
"normalized_example_id": "ex_002367",
"source_idx": 2367,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8751,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, k^2 + u)",
"budget": 5141
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002370",
"normalized_example_id": "ex_002370",
"source_idx": 2370,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8569,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, k^2 + u)",
"budget": 6855
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002370",
"normalized_example_id": "ex_002370",
"source_idx": 2370,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8569,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.