question stringlengths 64 859 | answer stringlengths 1 9 | 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 1
value | original_answer stringlengths 57 1.2k | 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Yolanda scored 345 points over the entire season. There were 15 games over the season. She averaged 4 free throws (worth one point), and 5 two-point baskets per game. How many three-point baskets did she average a game? | 3 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | She averaged 23 points per game because 345 / 15 = <<345/15=23>>23
She averaged 10 points a game off two-point baskets because 5 x 2 = 10
She averaged 4 points a game from free throws because 4 x 1 = <<4*1=4>>4
She averaged 14 points a game off non-three point baskets because 10 + 4 = <<10+4=14>>14
She averaged 9 point... | math | numeric_equivalence | gsm8k | main | gsm8k_1189 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nYolanda scored 345 points over the entire season. There were 15 games over the season. She averaged 4 free throws (worth one point) and 5 two-point baskets per game. How many three-point baskets did she average per game?\n\nStep 1: Calculate the total points from free throws.\n- Yolanda made... | [
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/Qw... | {
"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,
true,
false,
false,
false,
false,
false
] | [
null,
"13",
"22.07",
"14",
"(14)",
"(7.33)",
null,
null,
null,
"14",
"3",
"70",
"4.67",
null,
"70",
"13.2"
] | [
{
"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": "13",
"span_end": 1309,
"span_start": 1307... | [
{
"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
],
[
false
],
[
false
],
[
false
],
[],
[],
[],
[
false
],
[
true
],
[
false
],
[
false
],
[],
[
false
],
[
false
]
] | [
[],
[
"13"
],
[
"22.07"
],
[
"14"
],
[
"(14)"
],
[
"(7.33)"
],
[],
[],
[],
[
"14"
],
[
"3"
],
[
"70"
],
[
"4.67"
],
[],
[
"70"
],
[
"13.2"
]
] | [
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "13",
"pattern_used": "answer tag",
"position": 1307,
"span_end": 1309,
"span_start": 1307,
"span_text": "13"
}
],
[
{
"confidence": 1,
"extraction_type": "internal"... | [
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "3",
"evaluation_method": "math_equivalence",
"extracted_answer": "13",
"is_equivalent": false
}
],
[
{
"comparison_type": "mathematical",
"correct_answer": "3",
"evaluation_method": "math_... | {
"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_correct": 0.0625,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
... |
A basketball team played 40 games and won 70% of the games. It still had 10 games to play. How many games can they lose to win 60% of their games? | 8 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | The team has won 40 x 70/100 = <<40*70/100=28>>28 games out of 40.
There a total of 40 + 10 = <<40+10=50>>50 games to be played.
The team has to win a total of is 50 x 60/100 = <<50*60/100=30>>30 games.
So, it still needs to win 30 - 28 = <<30-28=2>>2 more games.
Hence, they can still lose 10 - 2 = <<10-2=8>>8 games.
#... | math | numeric_equivalence | gsm8k | main | gsm8k_1140 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nA basketball team has already played 40 games and won 70%. They have 10 more games left. What's the maximum number of losses they can have while still winning at least 60% of all their games?",
"Rewritten Question:\n\nA basketball team has already played 40 games and won 70% of them. They ... | [
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/Qw... | {
"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,
"(2)",
null,
null,
"25",
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
],
[],
[],
[],
[],
[]
] | [
[],
[],
[],
[],
[],
[],
[],
[
"(2)"
],
[],
[],
[
"25"
],
[],
[],
[],
[],
[]
] | [
[],
[],
[],
[],
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "(2)",
"pattern_used": "answer tag",
"position": 797,
"span_end": 802,
"span_start": 797,
"span_text": "(2)"
}
],
[],
[],
[
{
"conf... | [
[],
[],
[],
[],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "8",
"evaluation_method": "math_equivalence",
"extracted_answer": "(2)",
"is_equivalent": false
}
],
[],
[],
[
{
"comparison_type": "mathematical",
"correc... | {
"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,
... |
Fern is checking IDs to get into an R-rated movie. She denied 20% of the 120 kids from Riverside High, 70% of the 90 kids from West Side High, and half the 50 kids from Mountaintop High. How many kids got into the movie? | 148 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | First find how many kids from Riverside High are rejected: 20% * 120 kids = <<20*.01*120=24>>24 kids
Then find how many kids from West Side High are rejected: 70% * 90 kids = <<70*.01*90=63>>63 kids
Then find how many kids from Mountaintop High are rejected: 50 kids / 2 = <<50/2=25>>25 kids
Then add the number of kids ... | math | numeric_equivalence | gsm8k | main | gsm8k_140 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nFern denies IDs for an R-rated movie. She denies 20% of the 120 kids from Riverside High, 70% of the 90 kids from West Side High, and half the 50 kids from Mountaintop High. How many kids got into the movie?\n\nStep-by-step reasoning:\n1. Calculate the number of kids denied from Riverside Hi... | [
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/Qw... | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
true,
false,
true,
true,
true,
true,
true,
true,
true,
true,
false,
true,
false,
true,
true
] | [
null,
"148",
null,
"148",
"(148)",
"148",
"148",
"148",
"148",
"(148)",
"148",
null,
"148",
"112",
"148",
"148"
] | [
{
"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": "148",
"span_end": 950,
"span_start": 947,... | [
{
"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... | [
[],
[
true
],
[],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[],
[
true
],
[
false
],
[
true
],
[
true
]
] | [
[],
[
"148"
],
[],
[
"148"
],
[
"(148)"
],
[
"148"
],
[
"148"
],
[
"148"
],
[
"148"
],
[
"(148)"
],
[
"148"
],
[],
[
"148"
],
[
"112"
],
[
"148"
],
[
"148"
]
] | [
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "148",
"pattern_used": "answer tag",
"position": 947,
"span_end": 950,
"span_start": 947,
"span_text": "148"
}
],
[],
[
{
"confidence": 1,
"extraction_type": "inte... | [
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "148",
"evaluation_method": "math_equivalence",
"extracted_answer": "148",
"is_equivalent": true
}
],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "148",
"evaluation_metho... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 12,
"pass_at_n": 1,
"percent_correct": 0.75,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
... |
Cindy had 20 marbles which is 5 more than what Lisa had. If Cindy gave her 12 marbles, how many more marbles does Lisa have now? | 19 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | Lisa had 20 - 5 = <<20-5=15>>15 marbles.
Since Cindy gave 12 marbles, then she only has 20 - 12 = <<20-12=8>>8 marbles left.
Since Lisa received 12 marbles, then Lisa has 15 + 12 = 27 marbles now.
Hence, Lisa has 27 - 8 = <<27-8=19>>19 marbles more than Cindy.
#### 19 | math | numeric_equivalence | gsm8k | main | gsm8k_7055 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nCindy initially has 20 marbles. Lisa has fewer marbles than Cindy by exactly 5. How many more marbles does Lisa have after Cindy gives her 12 extra marbles?",
"Rewritten Question:\n\nCindy has 20 marbles that is 5 more than what Lisa has. After giving away 12 marbles to Cindy, how many mor... | [
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/Qw... | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
true,
false,
false,
false,
false,
false,
false,
false,
false,
false,
false
] | [
null,
null,
null,
"27",
null,
"19",
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
],
[],
[
true
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[]
] | [
[],
[],
[],
[
"27"
],
[],
[
"19"
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[]
] | [
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "27",
"pattern_used": "answer tag",
"position": 1012,
"span_end": 1014,
"span_start": 1012,
"span_text": "27"
}
],
[],
[
{
"confidence": 1,
"extraction... | [
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "19",
"evaluation_method": "math_equivalence",
"extracted_answer": "27",
"is_equivalent": false
}
],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "19",
"evalua... | {
"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_correct": 0.0625,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
... |
At Snowflake Plastics, each employee gets 10 sick days and 10 vacation days per year. If Mark uses half his allotment of both types of days in a year, how many hours' worth of days does he have left if each day covers an 8-hour long workday? | 80 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | First, we add the two types of days together to find 10+10=<<10+10=20>>20 days in total between the two types.
We then divide this number in half to find Mark's remaining number of days, getting 20/2= <<10=10>>10 days remaining.
Since each day counts as 8 hours, we multiply 10*8= <<10*8=80>>80 hours' worth of days rema... | math | numeric_equivalence | gsm8k | main | gsm8k_466 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nMark works at Snowflake Plastics and has been given 20 days off total (10 sick days + 10 vacation days) every year. He used up half of his sick days and half of his vacation days last year. How many hours' worth of days does he still have left? Each day covers an 8-hour workday.\n\nStep-by-s... | [
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/Qw... | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
true,
true,
false,
false,
false,
false,
false,
false,
true,
true,
false,
false,
false,
false,
false,
false
] | [
"(80)",
"80",
"160",
null,
"160",
null,
"160",
"160",
"80",
"80",
"10",
"0",
"160",
null,
"(160)",
null
] | [
{
"extraction_type": "final",
"non_overlapping_spans": 1,
"original_span": "(80)",
"span_end": 1059,
"span_start": 1053,
"total_spans_found": 1
},
{
"extraction_type": "final",
"non_overlapping_spans": 1,
"original_span": "80",
"span_end": 894,
"span_start": 890,
... | [
{
"comparison_type": "mathematical",
"correct_answer": "80",
"evaluation_method": "math_equivalence",
"extracted_answer": "(80)",
"extraction_result": null,
"is_equivalent": true,
"reason": null
},
{
"comparison_type": "mathematical",
"correct_answer": "80",
"evaluation_m... | [
[
true
],
[
true
],
[
false
],
[],
[
false
],
[],
[
false
],
[
false
],
[
true
],
[
true
],
[
false
],
[
false
],
[
false
],
[],
[
false
],
[]
] | [
[
"(80)"
],
[
"80"
],
[
"160"
],
[],
[
"160"
],
[],
[
"160"
],
[
"160"
],
[
"80"
],
[
"80"
],
[
"10"
],
[
"0"
],
[
"160"
],
[],
[
"(160)"
],
[]
] | [
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "(80)",
"pattern_used": "answer tag",
"position": 1053,
"span_end": 1059,
"span_start": 1053,
"span_text": "(80)"
}
],
[
{
"confidence": 1,
"extraction_type": "internal",
... | [
[
{
"comparison_type": "mathematical",
"correct_answer": "80",
"evaluation_method": "math_equivalence",
"extracted_answer": "(80)",
"is_equivalent": true
}
],
[
{
"comparison_type": "mathematical",
"correct_answer": "80",
"evaluation_method": "math_equ... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 4,
"pass_at_n": 1,
"percent_correct": 0.25,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
... |
John books 3 nights at a hotel room for $250 a night. He has a $100 discount. How much does he pay? | 650 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | The cost is 250*3=$<<250*3=750>>750
So he spends 750-100=$<<750-100=650>>650
#### 650 | math | numeric_equivalence | gsm8k | main | gsm8k_2471 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nJohn stays at a hotel for 3 nights. Each night costs $250. He gets a $100 discount. How much does John pay?",
"Rewritten Question:\n\nHow much does John pay after applying his discount on a three-night stay at a hotel where each night costs $250? \n\nStep-by-step thinking:\n\n1. Calculate ... | [
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/Qw... | {
"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,
"$650",
null,
null,
null,
null,
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": 1,
"original_span": "$650",
"span_end": 471,
"span_start": 465... | [
{
"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
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[]
] | [
[],
[
"$650"
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[]
] | [
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "$650",
"pattern_used": "answer tag",
"position": 465,
"span_end": 471,
"span_start": 465,
"span_text": "$650"
}
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
... | [
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "650",
"evaluation_method": "math_equivalence",
"extracted_answer": "$650",
"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,
... |
A library cabinet houses five ancient scrolls. The first scroll is 4080 years old. If each scroll is older than the last by half as many years as the last scroll’s age, how old is the fifth scroll? | 20655 | gsm8k | gsm8k | [
{
"content": "You answer questions by saying \"Rewritten Question: \" then rewriting the question to only contain what is needed to solve it and then think step by step and then you give your final answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving... | [] | [] | rephrase | The second scroll is 4080 + 4080 / 2 = 4000 + 2040 = <<4080+4080/2=6120>>6120 years old.
The third scroll is 6120 + 6120 / 2 = 6120 + 3060 = <<6120+6120/2=9180>>9180 years old.
The fourth scroll is 9180 + 9180 / 2 = 9180 + 4590 = <<9180+9180/2=13770>>13770 years old.
The fifth scroll is 13770 + 13770 / 2 = 13770 + 6885... | math | numeric_equivalence | gsm8k | main | gsm8k_224 | boxed_number | elementary_math | arithmetic_word_problems | [
"Rewritten Question:\n\nA library cabinet houses five ancient scrolls. The first scroll is 4080 years old. If each scroll is older than the last by half as many years as the last scroll's age, how old is the fifth scroll?",
"Rewritten Question:\n\nThe first scroll is 4080 years old. Each subsequent scroll is olde... | [
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/Qw... | {
"backend": "litellm",
"model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct",
"n_responses": 16
} | [
false,
false,
false,
false,
false,
false,
false,
true,
true,
false,
false,
false,
false,
true,
false,
true
] | [
null,
null,
null,
"0",
null,
null,
null,
"20655",
"20655",
null,
null,
"255",
"10880",
"20655",
null,
"20655"
] | [
{
"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
],
[],
[],
[],
[
true
],
[
true
],
[],
[],
[
false
],
[
false
],
[
true
],
[],
[
true
]
] | [
[],
[],
[],
[
"0"
],
[],
[],
[],
[
"20655"
],
[
"20655"
],
[],
[],
[
"255"
],
[
"10880"
],
[
"20655"
],
[],
[
"20655"
]
] | [
[],
[],
[],
[
{
"confidence": 1,
"extraction_type": "internal",
"original_span": "0",
"pattern_used": "answer tag",
"position": 2370,
"span_end": 2371,
"span_start": 2370,
"span_text": "0"
}
],
[],
[],
[],
[
{
"confidence": 1,
"... | [
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "20655",
"evaluation_method": "math_equivalence",
"extracted_answer": "0",
"is_equivalent": false
}
],
[],
[],
[],
[
{
"comparison_type": "mathematical",
"correct_answer": "2065... | {
"flips_by": [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
],
"flips_total": 0,
"num_correct": 4,
"pass_at_n": 1,
"percent_correct": 0.25,
"skill_count": {
"answer_revision": [
0,
0,
0,
0,
0,
0,
... |
"On Tuesday, Mike had 45 books and Corey had twice as many books as Mike. On Wednesday, Mike gave 10(...TRUNCATED) | 35 | gsm8k | gsm8k | [{"content":"You answer questions by saying \"Rewritten Question: \" then rewriting the question to (...TRUNCATED) | [] | [] | rephrase | "Corey gave Lily 10+15 = <<10+15=25>>25 books to Lily.\nLily got 25+10 = <<25+10=35>>35 books.\n####(...TRUNCATED) | math | numeric_equivalence | gsm8k | main | gsm8k_3278 | boxed_number | elementary_math | arithmetic_word_problems | ["Rewritten Question:\n\nMike has 45 books on Tuesday. Corey has twice as many books as Mike. On Wed(...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,
true,
true,
true,
false,
true,
true,
false,
false,
false,
true,
false
] | [
null,
null,
null,
null,
null,
"35",
"35",
"35",
null,
"35",
"35",
null,
null,
null,
"35",
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) | [
[],
[],
[],
[],
[],
[
true
],
[
true
],
[
true
],
[],
[
true
],
[
true
],
[],
[],
[],
[
true
],
[]
] | [
[],
[],
[],
[],
[],
[
"35"
],
[
"35"
],
[
"35"
],
[],
[
"35"
],
[
"35"
],
[],
[],
[],
[
"35"
],
[]
] | [[],[],[],[],[],[{"confidence":1.0,"extraction_type":"internal","original_span":"35","pattern_used":(...TRUNCATED) | [[],[],[],[],[],[{"comparison_type":"mathematical","correct_answer":"35","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":6,"pass_at_n":1,"percent(...TRUNCATED) |
"John has five more roommates than twice as many as Bob. If Bob has 10 roommates, how many roommates(...TRUNCATED) | 25 | gsm8k | gsm8k | [{"content":"You answer questions by saying \"Rewritten Question: \" then rewriting the question to (...TRUNCATED) | [] | [] | rephrase | "Twice ten roommates is 2*10 = <<2*10=20>>20\nJohn has 5 more than 20 roommates which is 20+5 = <<5+(...TRUNCATED) | math | numeric_equivalence | gsm8k | main | gsm8k_126 | boxed_number | elementary_math | arithmetic_word_problems | ["Rewritten Question:\n\nIf Bob has 10 roommates, how many roommates does John have if John has five(...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
} | [
true,
true,
true,
true,
true,
true,
true,
true,
true,
true,
false,
false,
false,
false,
false,
false
] | [
"25",
"25",
"25",
"25",
"(25)",
"25",
"25",
"25",
"25",
"25",
null,
null,
null,
null,
null,
null
] | [{"extraction_type":"final","non_overlapping_spans":1,"original_span":"25","span_end":438,"span_star(...TRUNCATED) | [{"comparison_type":"mathematical","correct_answer":"25","evaluation_method":"math_equivalence","ext(...TRUNCATED) | [
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[
true
],
[],
[],
[],
[],
[],
[]
] | [
[
"25"
],
[
"25"
],
[
"25"
],
[
"25"
],
[
"(25)"
],
[
"25"
],
[
"25"
],
[
"25"
],
[
"25"
],
[
"25"
],
[],
[],
[],
[],
[],
[]
] | [[{"confidence":1.0,"extraction_type":"internal","original_span":"25","pattern_used":"answer tag","p(...TRUNCATED) | [[{"comparison_type":"mathematical","correct_answer":"25","evaluation_method":"math_equivalence","ex(...TRUNCATED) | {"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":10,"pass_at_n":1,"percen(...TRUNCATED) |
"Caleb, Andy and Billy went on a picnic with their father. Billy took 6 candies with him, Caleb took(...TRUNCATED) | 4 | gsm8k | gsm8k | [{"content":"You answer questions by saying \"Rewritten Question: \" then rewriting the question to (...TRUNCATED) | [] | [] | rephrase | "Billy took some candies and got more from his father for a total of 6+8 = <<6+8=14>>14 candies\nCal(...TRUNCATED) | math | numeric_equivalence | gsm8k | main | gsm8k_1836 | boxed_number | elementary_math | arithmetic_word_problems | ["Rewritten Question:\n\nBilly had 6 candies initially. Caleb had 11 candies. Andy had 9 candies. Th(...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
] | [
"6",
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
null,
"6",
null,
null
] | [{"extraction_type":"final","non_overlapping_spans":1,"original_span":"6","span_end":825,"span_start(...TRUNCATED) | [{"comparison_type":"mathematical","correct_answer":"4","evaluation_method":"math_equivalence","extr(...TRUNCATED) | [
[
false
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
false
],
[],
[]
] | [
[
"6"
],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[],
[
"6"
],
[],
[]
] | [[{"confidence":1.0,"extraction_type":"internal","original_span":"6","pattern_used":"answer tag","po(...TRUNCATED) | [[{"comparison_type":"mathematical","correct_answer":"4","evaluation_method":"math_equivalence","ext(...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) |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 4