question stringlengths 51 985 | answer stringlengths 1 11 | task_config stringclasses 1
value | task_source stringclasses 1
value | prompt listlengths 2 2 | model_responses listlengths 0 0 | model_responses__eval_is_correct listlengths 0 0 | prompt_variant stringclasses 2
values | original_answer stringlengths 51 1.12k | task_type stringclasses 1
value | evaluation_type stringclasses 1
value | source stringclasses 1
value | variant stringclasses 1
value | id stringlengths 7 10 | expected_answer_format stringclasses 1
value | difficulty stringclasses 1
value | domain stringclasses 1
value | model_responses__mutated_prompts listlengths 16 16 | model_responses__mutated_prompts__finish_reason_length_flags listlengths 16 16 | model_responses__mutated_prompts__length_partial_responses listlengths 16 16 | prompt__mutated_prompts__metadata dict | model_responses__mutated_prompts__metadata dict | model_responses__mutated_prompts__eval_is_correct listlengths 16 16 | model_responses__mutated_prompts__eval_extracted_answers listlengths 16 16 | model_responses__mutated_prompts__eval_extraction_metadata listlengths 16 16 | model_responses__mutated_prompts__eval_evaluation_metadata listlengths 16 16 | model_responses__mutated_prompts__internal_answers__eval_is_correct listlengths 16 16 ⌀ | model_responses__mutated_prompts__internal_answers__eval_extracted_answers listlengths 16 16 ⌀ | model_responses__mutated_prompts__internal_answers__eval_extraction_metadata listlengths 16 16 ⌀ | model_responses__mutated_prompts__internal_answers__eval_evaluation_metadata listlengths 16 16 ⌀ | model_responses__mutated_prompts__metrics dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
A colony of bees can contain up to 80000 individuals. In winter they are more exposed to death, and if the winter is really cold the bees can begin to die slowly. If the colony starts to lose 1200 bees per day, how many days will pass until the number of bees in the colony reaches a fourth of its initial number? | 50 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | A fourth of the initial number of bees is 80000 bees / 4 = <<80000/4=20000>>20000 bees
The number of bees that has to die is 80000 bees – 20000 bees = <<80000-20000=60000>>60000 bees
For the colony to reach a fourth of its population, 60000 bees ÷ 1200 bees/day = <<60000/1200=50>>50 days have to pass.
#### 50 | math | numeric_equivalence | gsm8k | main | gsm8k_2457 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n# Define constants\r\nMAX_BEE_COUNT = 80000\r\nBEES_LOST_PER_DAY = 1200\r\n\r\n# Calculate the target bee count (fourth of maximum)\r\nTARGET_BEES = MAX_BEE_COUNT / 4\r\n\r\n# Calculate the number of days it takes for the colony to reach the target bee count\r\nDAYS_TO_REACH_TARGET = TARGET_BEES / BEES_... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nA colony of bees can contain up to 80000 individuals. In winter they are more exposed to death, and if the winter is really cold the bees can begin to die slowly. If the colony starts to lose 1200 bees per day, how many days will pass until the number of bees in the colony reaches a fourth of its initial number?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
"63",
null,
null,
null,
null,
null,
null,
null,
null,
null,
"375",
null,
null,
null,
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": 1,
"original_span": "63",
"span_end": 654,
"span_start": 652,
... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": "mathematical",
"correct... | [
[],
[
false
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
false
],
[],
[],
[],
[]
] | [
[],
[
"63"
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
"375"
],
[],
[],
[],
[]
] | [
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "63",
"pattern_used": "answer tag",
"position": 652,
"span_end": 654,
"span_start": 652,
"span_text": "63"
}
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
{
"... | [
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "50",
"evaluation_method": "math_equivalence",
"extracted_answer": "63",
"is_equivalent": false
}
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
Tina is a professional boxer. She wins her first 10 fights of her career. She then goes on to win 5 more before losing her first fight, and then doubles her number of wins before losing again. She then retires. How many more wins than losses does she have at the end of her career? | 28 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | Tina wins her first 10, and then wins 5 more for 10+5 = 15 wins before losing 1.
She then doubles her number of wins for a total of 15*2=30 wins, before losing 1 more for 1+1=2 losses.
For her career she has 30-2=<<30-2=28>>28 more wins than losses.
#### 28 | math | numeric_equivalence | gsm8k | main | gsm8k_3372 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n\n# Tina's initial wins after the first 10 fights\ninitial_wins = 10\n\n# Number of additional wins before losing her first fight\nadditional_wins_before_first_loss = 5\n\n# Wins doubled after losing her first fight\nwins_doubled_after_first_loss = additional_wins_before_first_loss * 2\n\n# Total wins b... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nTina is a professional boxer. She wins her first 10 fights of her career. She then goes on to win 5 more before losing her first fight, and then doubles her number of wins before losing again. She then retires. How many more wins than losses does she have at the end of her career?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
null,
null,
null,
"-1",
"-1",
null,
null,
null,
null,
null,
"6",
null,
null,
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": n... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": null,
"correct_answer": ... | [
[],
[],
[],
[],
[],
[
false
],
[
false
],
[],
[],
[],
[],
[],
[
false
],
[],
[],
[]
] | [
[],
[],
[],
[],
[],
[
"-1"
],
[
"-1"
],
[],
[],
[],
[],
[],
[
"6"
],
[],
[],
[]
] | [
[],
[],
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "-1",
"pattern_used": "answer tag",
"position": 971,
"span_end": 973,
"span_start": 971,
"span_text": "-1"
}
],
[
{
"confidence": 1,
"extract... | [
[],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "28",
"evaluation_method": "math_equivalence",
"extracted_answer": "-1",
"is_equivalent": false
}
],
[
{
"comparison_type": "mathematical",
"correct_answer": "28",
"... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
Tracy has a charity book collection store from which she receives book donations from members of her town. Any person who wants to borrow a book is free to do so as long as they return the book after they've finished reading. She has already received 300 book donations. If ten people come and donate five books each in a week, and a total of 140 books are borrowed by other people to read, calculate the total number of books she has remaining in the store? | 210 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | The ten people donated 10*5 = <<10*5=50>>50 books.
The total number of books in her store after the new donations is 300+50 = <<300+50=350>>350 books.
When some of the books are borrowed, her total collection becomes 350-140 =<<350-140=210>>210 books.
#### 210 | math | numeric_equivalence | gsm8k | main | gsm8k_3077 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n\n# Initial number of book donations Tracy has\ninitial_donations = 300\n\n# Number of people donating books in a week\ndonors_per_week = 10\n\n# Number of books donated per donor\nbooks_per_donor = 5\n\n# Total number of books borrowed\nborrowed_books = 140\n\n# Calculate the total number of books dona... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nTracy has a charity book collection store from which she receives book donations from members of her town. Any person who wants to borrow a book is free to do so as long as they return the book after they've finished reading. She has already received 300 book donations. If ten people come and donate five books each in a week, and a total of 140 books are borrowed by other people to read, calculate the total number of books she has remaining in the store?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
null,
null,
null,
null,
"295",
null,
"800",
null,
null,
null,
null,
null,
null,
"600"
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": n... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": null,
"correct_answer": ... | [
[],
[],
[],
[],
[],
[],
[
false
],
[],
[
false
],
[],
[],
[],
[],
[],
[],
[
false
]
] | [
[],
[],
[],
[],
[],
[],
[
"295"
],
[],
[
"800"
],
[],
[],
[],
[],
[],
[],
[
"600"
]
] | [
[],
[],
[],
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "295",
"pattern_used": "answer tag",
"position": 463,
"span_end": 466,
"span_start": 463,
"span_text": "295"
}
],
[],
[
{
"confidence": 1,
... | [
[],
[],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "210",
"evaluation_method": "math_equivalence",
"extracted_answer": "295",
"is_equivalent": false
}
],
[],
[
{
"comparison_type": "mathematical",
"correct_answer":... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
Lionel walked 4 miles. Esther walked 975 yards and Niklaus walked 1287 feet. How many combined feet did the friends walk? | 25332 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | Lionel = 4 * 5280 = <<4*5280=21120>>21120 feet
Esther = 975 * 3 = <<975*3=2925>>2925 feet
Together = 21120 + 2925 + 1287 = <<21120+2925+1287=25332>>25332 feet
Together the friends walked 25332 feet.
#### 25332 | math | numeric_equivalence | gsm8k | main | gsm8k_2838 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n\n# Convert distances from different units to feet\nlionel_distance_feet = 4 * 3.28084 # Lionel's distance in feet\nesther_distance_feet = 975 / 3.28084 # Esther's distance in feet\nniklaus_distance_feet = 1287 / 3.28084 # Niklaus's distance in feet\n\n# Calculate total distance in feet\ntotal_distan... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nLionel walked 4 miles. Esther walked 975 yards and Niklaus walked 1287 feet. How many combined feet did the friends walk?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
"(26403)",
null,
null,
"62790",
null,
null,
null,
null,
null,
null,
null,
null,
null,
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": n... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": null,
"correct_answer": ... | [
[],
[],
[
false
],
[],
[],
[
false
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[]
] | [
[],
[],
[
"(26403)"
],
[],
[],
[
"62790"
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[]
] | [
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "(26403)",
"pattern_used": "answer tag",
"position": 576,
"span_end": 585,
"span_start": 576,
"span_text": "(26403)"
}
],
[],
[],
[
{
"confidence": 1,
"ext... | [
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "25332",
"evaluation_method": "math_equivalence",
"extracted_answer": "(26403)",
"is_equivalent": false
}
],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "25332",
... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
There are 60 pieces of chips in Lara's bag. One-sixth of the chips are blue. There are 34 red chips and the rest are green. How many green chips are in Lara's bag? | 16 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | There are 60 x 1/6 = <<60*1/6=10>>10 blue chips.
There are 10 + 34 = <<10+34=44>>44 chips that are not green.
Thus, 60 - 44 = <<60-44=16>>16 chips are green.
#### 16 | math | numeric_equivalence | gsm8k | main | gsm8k_2095 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n\n# Total number of chips\ntotal_chips = 60\n\n# Number of blue chips (one-sixth of total chips)\nblue_chips = total_chips / 6\n\n# Number of red chips\nred_chips = 34\n\n# Calculate the number of green chips\ngreen_chips = total_chips - blue_chips - red_chips\n\ngreen_chips\n```",
"```python\n# Total... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nThere are 60 pieces of chips in Lara's bag. One-sixth of the chips are blue. There are 34 red chips and the rest are green. How many green chips are in Lara's bag?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
null,
null,
null,
null,
null,
null,
"12",
null,
null,
"12",
null,
null,
"(18)",
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": n... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": null,
"correct_answer": ... | [
[],
[],
[],
[],
[],
[],
[],
[],
[
false
],
[],
[],
[
false
],
[],
[],
[
false
],
[]
] | [
[],
[],
[],
[],
[],
[],
[],
[],
[
"12"
],
[],
[],
[
"12"
],
[],
[],
[
"(18)"
],
[]
] | [
[],
[],
[],
[],
[],
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "12",
"pattern_used": "answer tag",
"position": 362,
"span_end": 364,
"span_start": 362,
"span_text": "12"
}
],
[],
[],
[
{
"... | [
[],
[],
[],
[],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "16",
"evaluation_method": "math_equivalence",
"extracted_answer": "12",
"is_equivalent": false
}
],
[],
[],
[
{
"comparison_type": "mathematical",
"... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
Jessica has one hour to take an exam. She has answered 16 out of 80 questions. She has used 12 minutes of her time. If she keeps up this same pace, how many minutes will be left when she finishes the exam? | 0 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | She takes .75 minutes to answer a question because 12 / 16 = <<12/16=.75>>.75
It will take her 60 minutes to take the entire exam because 80 x .75 = <<80*.75=60>>60
She has 0 minutes left when she is done because 60 - 60 = <<60-60=0>>0
#### 0 | math | numeric_equivalence | gsm8k | main | gsm8k_626 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n# Constants\ntotal_questions = 80\ntime_used_minutes = 12\n\n# Calculate remaining time per question\nremaining_time_per_question = total_time - (questions_answered * time_per_question)\n\n# Calculate total remaining time for all questions\ntotal_remaining_time = remaining_time_per_question * questions_... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nJessica has one hour to take an exam. She has answered 16 out of 80 questions. She has used 12 minutes of her time. If she keeps up this same pace, how many minutes will be left when she finishes the exam?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
"34",
null,
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": n... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": null,
"correct_answer": ... | [
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
false
],
[],
[]
] | [
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
"34"
],
[],
[]
] | [
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "34",
"pattern_used": "answer tag",
"position": 852,
"span_end": 854,
"span_start": 852,
"span_text": "34"
}
],... | [
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "0",
"evaluation_method": "math_equivalence",
"extracted_answer": "34",
"is_equivalent": false
}
],
[],
[]
] | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
There are 48 crayons in the box. Kiley takes 1/4 of them away. Joe takes away half of the remaining crayons, how many crayons are left? | 18 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | Kiley takes 48/4=<<48/4=12>>12 crayons
48-12=<<48-12=36>>36 crayons remain
Joe takes 36/2=<<36/2=18>>18 crayons
36-18=<<36-18=18>>18 crayons left
#### 18 | math | numeric_equivalence | gsm8k | main | gsm8k_6433 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n# Initial number of crayons\ninitial_crayons = 48\n\n# Kiley takes 1/4 of them away\ncrayons_taken_by_kiley = initial_crayons / 4\nremaining_after_kiley = initial_crayons - crayons_taken_by_kiley\n\n# Joe takes half of the remaining crayons\ncrayons_taken_by_joe = remaining_after_kiley / 2\nfinal_remain... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nThere are 48 crayons in the box. Kiley takes 1/4 of them away. Joe takes away half of the remaining crayons, how many crayons are left?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
"9",
null,
null,
null,
null,
"36",
null,
null,
"27",
null,
null,
null,
null,
"12"
] | [
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
"total_spans_found": null
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": n... | [
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "response_processor",
"extracted_answer": null,
"extraction_result": "{'extraction_type': 'final'}",
"is_equivalent": null,
"reason": "no_extracted_answer"
},
{
"comparison_type": null,
"correct_answer": ... | [
[],
[],
[
false
],
[],
[],
[],
[],
[
false
],
[],
[],
[
false
],
[],
[],
[],
[],
[
false
]
] | [
[],
[],
[
"9"
],
[],
[],
[],
[],
[
"36"
],
[],
[],
[
"27"
],
[],
[],
[],
[],
[
"12"
]
] | [
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "9",
"pattern_used": "answer tag",
"position": 400,
"span_end": 401,
"span_start": 400,
"span_text": "9"
}
],
[],
[],
[],
[],
[
{
"confidence": 1,
"ext... | [
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "18",
"evaluation_method": "math_equivalence",
"extracted_answer": "9",
"is_equivalent": false
}
],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "18",
... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
Nellie had 380 legos, but she lost 57 of them and gave her sister 24 legos. How many legos does she have now? | 299 | gsm8k | gsm8k | [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "An... | [] | [] | python_prog | After losing 57 legos, Nellie has 380 - 57 = <<380-57=323>>323 legos.
After giving her sister 24 legos Nellie has 323 - 24 = <<323-24=299>>299 legos
#### 299 | math | numeric_equivalence | gsm8k | main | gsm8k_2466 | boxed_number | elementary_math | arithmetic_word_problems | [
"```python\n# Initial number of legos\ninitial_legos = 380\n\n# Number of legos lost\nlost_legos = 57\n\n# Number of legos given to sister\ngiven_legos = 24\n\n# Calculate remaining legos after losing some and giving some away\nremaining_legos = initial_legos - lost_legos - given_legos\n\n# Output the result\nprint... | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {
"api_url": "http://127.0.0.1:45035",
"backend": "litellm",
"chat_template_applied": true,
"generation_params": {
"chat_template_applied": true,
"max_tokens": 4096,
"n": 16,
"repetition_penalty": 1.1,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8
},
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"prompt": [
{
"content": "You like to solve problems by thinking in python code. Write a python script that would solve the question then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.",
"role": "system"
},
{
"content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nNellie had 380 legos, but she lost 57 of them and gave her sister 24 legos. How many legos does she have now?\n\nGive your answer in the following format:\n<answer>\n(your answer)\n</answer>\n\nWhere \"(your answer)\" is the numerical solution to the problem.\n\nTo solve this question, begin your thinking with \"```python\n\" write out the python program, then give your resulting number as the answer in the \"<answer>\n(your answer)\n</answer>\" tag.",
"role": "user"
}
]
} | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"109",
null,
null,
"109",
null,
null,
"109",
null,
"109",
null,
"109",
"101",
null,
"109",
null,
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": 1,
"original_span": "109",
"span_end": 376,
"span_start": 373,
"total_spans_found": 1
},
{
"extraction_type": "final",
"non_overlapping_spans": null,
"original_span": null,
"span_end": null,
"span_start": null,
... | [
{
"comparison_type": "mathematical",
"correct_answer": "299",
"evaluation_method": "math_equivalence",
"extracted_answer": "109",
"extraction_result": null,
"is_equivalent": false,
"reason": null
},
{
"comparison_type": null,
"correct_answer": null,
"evaluation_method": "... | [
[
false
],
[],
[],
[
false
],
[],
[],
[
false
],
[],
[
false
],
[],
[
false
],
[
false
],
[],
[
false
],
[],
[]
] | [
[
"109"
],
[],
[],
[
"109"
],
[],
[],
[
"109"
],
[],
[
"109"
],
[],
[
"109"
],
[
"101"
],
[],
[
"109"
],
[],
[]
] | [
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "109",
"pattern_used": "answer tag",
"position": 373,
"span_end": 376,
"span_start": 373,
"span_text": "109"
}
],
[],
[],
[
{
"confidence": 1,
"extraction_type": "inte... | [
[
{
"comparison_type": "mathematical",
"correct_answer": "299",
"evaluation_method": "math_equivalence",
"extracted_answer": "109",
"is_equivalent": false
}
],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "299",
"evaluation_meth... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 0,
"pass_at_n": 0,
"percent_correct": 0,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"best_of_n": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflect_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"reflection_sbon": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"sample_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_close": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"vote_open": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"voting": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
]
},
"total_responses": 16
} |
"An alien invades Earth. It abducts 200 people. He returns 80% of the people abducted. After that(...TRUNCATED) | 30 | gsm8k | gsm8k | [{"content":"You like to solve problems by thinking in python code. Write a python script that would(...TRUNCATED) | [] | [] | python_prog | "He returned 200*.8=<<200*.8=160>>160 people\nSo he had 200-160=<<200-160=40>>40 people\nHe took 10 (...TRUNCATED) | math | numeric_equivalence | gsm8k | main | gsm8k_4523 | boxed_number | elementary_math | arithmetic_word_problems | ["```python\n# Initial abduction count\nabducted_people = 200\n\n# Return percentage\nreturn_percent(...TRUNCATED) | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {"api_url":"http://127.0.0.1:45035","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED) | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
null,
null,
null,
"50",
"40",
null,
null,
null,
null,
null,
"100",
"100.0",
null,
null
] | [{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED) | [{"comparison_type":null,"correct_answer":null,"evaluation_method":"response_processor","extracted_a(...TRUNCATED) | [
[],
[],
[],
[],
[],
[
false
],
[
false
],
[],
[],
[],
[],
[],
[
false
],
[
false
],
[],
[]
] | [
[],
[],
[],
[],
[],
[
"50"
],
[
"40"
],
[],
[],
[],
[],
[],
[
"100"
],
[
"100.0"
],
[],
[]
] | [[],[],[],[],[],[{"confidence":1.0,"extraction_type":"internal","original_span":"50","pattern_used":(...TRUNCATED) | [[],[],[],[],[],[{"comparison_type":"mathematical","correct_answer":"30","evaluation_method":"math_e(...TRUNCATED) | {"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent(...TRUNCATED) |
"Randy feeds his pigs 10 pounds of feed per pig per day. If Randy has 2 pigs, how many pounds of pi(...TRUNCATED) | 140 | gsm8k | gsm8k | [{"content":"You like to solve problems by thinking in python code. Write a python script that would(...TRUNCATED) | [] | [] | python_prog | "2 pigs fed for 7 days is a total of 7*2=<<7*2=14>>14 feedings.\nAt 10 pounds per feeding, this amou(...TRUNCATED) | math | numeric_equivalence | gsm8k | main | gsm8k_4035 | boxed_number | elementary_math | arithmetic_word_problems | ["```python\n# Given data\nfeed_per_pig_per_day = 10 # in pounds\nnumber_of_pigs = 2\n\n# Calculate(...TRUNCATED) | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
"",
""
] | {"api_url":"http://127.0.0.1:45035","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED) | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false,
true,
false,
false,
false
] | [
null,
null,
null,
null,
"60",
"(350)",
null,
null,
null,
"35",
null,
"350",
"140",
null,
"560",
null
] | [{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED) | [{"comparison_type":null,"correct_answer":null,"evaluation_method":"response_processor","extracted_a(...TRUNCATED) | [
[],
[],
[],
[],
[
false
],
[
false
],
[],
[],
[],
[
false
],
[],
[
false
],
[
true
],
[],
[
false
],
[]
] | [
[],
[],
[],
[],
[
"60"
],
[
"(350)"
],
[],
[],
[],
[
"35"
],
[],
[
"350"
],
[
"140"
],
[],
[
"560"
],
[]
] | [[],[],[],[],[{"confidence":1.0,"extraction_type":"internal","original_span":"60","pattern_used":"an(...TRUNCATED) | [[],[],[],[],[{"comparison_type":"mathematical","correct_answer":"140","evaluation_method":"math_equ(...TRUNCATED) | {"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":1,"pass_at_n":1,"percent(...TRUNCATED) |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 3