id stringlengths 7 41 | benchmark stringclasses 6
values | task_name stringclasses 6
values | source dict | prompt stringlengths 87 6.14k | reference_answer stringlengths 1 932 | reference_solution stringlengths 0 3.36k | source_record_json stringlengths 94 30.4k | prompt_format stringclasses 1
value | fewshot_count int32 0 5 | generation dict | assessment dict |
|---|---|---|---|---|---|---|---|---|---|---|---|
gsm8k:0 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 0
} | <|im_start|>user
Question: A housewife goes to the market. She spent 2/3 of her $150. How much does she have left?
Answer: The housewife has 3/3 - 2/3 = 1/3 of $150 left.
This means she has $150 * (1/3) = $<<150*(1/3)=50>>50 left.
#### 50
Question: Bart makes a mixtape. The first side has 6 songs. The second side ha... | Janet sells 16 - 3 - 4 = <<16-3-4=9>>9 duck eggs a day.
She makes 9 * 2 = $<<9*2=18>>18 every day at the farmer’s market.
#### 18 | {"answer": "Janet sells 16 - 3 - 4 = <<16-3-4=9>>9 duck eggs a day.\nShe makes 9 * 2 = $<<9*2=18>>18 every day at the farmer’s market.\n#### 18", "question": "Janet’s ducks lay 16 eggs per day. She eats three for breakfast every morning and bakes muffins for her friends every day with four. She sells the remainder at t... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:1 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 1
} | <|im_start|>user
Question: Hannah wants to save $80 for five weeks. In the first week, she saved $4 and she plans to save twice as much as her savings as the previous week. How much will she save in the fifth week to reach her goal?
Answer: In the second week, Hannah will save $4 x 2 = $<<4*2=8>>8.
In the third week, s... | It takes 2/2=<<2/2=1>>1 bolt of white fiber
So the total amount of fabric is 2+1=<<2+1=3>>3 bolts of fabric
#### 3 | {"answer": "It takes 2/2=<<2/2=1>>1 bolt of white fiber\nSo the total amount of fabric is 2+1=<<2+1=3>>3 bolts of fabric\n#### 3", "question": "A robe takes 2 bolts of blue fiber and half that much white fiber. How many bolts in total does it take?"} | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:2 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 2
} | <|im_start|>user
Question: Mel is three years younger than Katherine. When Katherine is two dozen years old, how old will Mel be in years?
Answer: When Katherine is 2 dozen years old, she will be 2*12=<<2*12=24>>24 years old.
If Mel is three years younger than Katherine, then when Katherine is 24 years old, Mel will b... | The cost of the house and repairs came out to 80,000+50,000=$<<80000+50000=130000>>130,000
He increased the value of the house by 80,000*1.5=<<80000*1.5=120000>>120,000
So the new value of the house is 120,000+80,000=$<<120000+80000=200000>>200,000
So he made a profit of 200,000-130,000=$<<200000-130000=70000>>70,000
#... | {"answer": "The cost of the house and repairs came out to 80,000+50,000=$<<80000+50000=130000>>130,000\nHe increased the value of the house by 80,000*1.5=<<80000*1.5=120000>>120,000\nSo the new value of the house is 120,000+80,000=$<<120000+80000=200000>>200,000\nSo he made a profit of 200,000-130,000=$<<200000-130000=... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:3 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 3
} | <|im_start|>user
Question: Rick has just celebrated his 15th birthday. His oldest brother is twice his age, and his middle brother is a third of the oldest brother’s age. The smallest brother is half the middle brother’s age. If the youngest brother is 2 years younger than the smallest brother, how old, in years, is th... | He sprints 3*3=<<3*3=9>>9 times
So he runs 9*60=<<9*60=540>>540 meters
#### 540 | {"answer": "He sprints 3*3=<<3*3=9>>9 times\nSo he runs 9*60=<<9*60=540>>540 meters\n#### 540", "question": "James decides to run 3 sprints 3 times a week. He runs 60 meters each sprint. How many total meters does he run a week?"} | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:4 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 4
} | <|im_start|>user
Question: A bookseller sells 15 books in January, 16 in February and a certain number of books in March. If the average number of books he sold per month across all three months is 16, how many books did he sell in March?
Answer: He sold an average of 16 books per month across all 3 months so he must h... | If each chicken eats 3 cups of feed per day, then for 20 chickens they would need 3*20=<<3*20=60>>60 cups of feed per day.
If she feeds the flock 15 cups of feed in the morning, and 25 cups in the afternoon, then the final meal would require 60-15-25=<<60-15-25=20>>20 cups of chicken feed.
#### 20 | {"answer": "If each chicken eats 3 cups of feed per day, then for 20 chickens they would need 3*20=<<3*20=60>>60 cups of feed per day.\nIf she feeds the flock 15 cups of feed in the morning, and 25 cups in the afternoon, then the final meal would require 60-15-25=<<60-15-25=20>>20 cups of chicken feed.\n#### 20", "ques... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:5 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 5
} | <|im_start|>user
Question: Jeremy decides to take Tylenol because of the pain he is experiencing. He decides to take 1000 mg every 6 hours for 2 weeks. How many pills do they take if they are each 500 mg?
Answer: He takes 1000/500=<<1000/500=2>>2 pills each time
He takes them 24/6=<<24/6=4>>4 times a day
So he takes 4... | The discount price of one glass is 60/100 * 5 = $<<60/100*5=3>>3.
If every second glass is cheaper, that means Kylar is going to buy 16 / 2 = <<16/2=8>>8 cheaper glasses.
So for the cheaper glasses, Kylar is going to pay 8 * 3 = $<<8*3=24>>24.
And for the regular-priced glasses, Kylar will pay 8 * 5 = $<<8*5=40>>40.
So... | {"answer": "The discount price of one glass is 60/100 * 5 = $<<60/100*5=3>>3.\nIf every second glass is cheaper, that means Kylar is going to buy 16 / 2 = <<16/2=8>>8 cheaper glasses.\nSo for the cheaper glasses, Kylar is going to pay 8 * 3 = $<<8*3=24>>24.\nAnd for the regular-priced glasses, Kylar will pay 8 * 5 = $<... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:6 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 6
} | <|im_start|>user
Question: Three frogs are trying to hop across the road. The first frog takes 4 times as many hops as the second. The second frog takes twice as many as the third. If the three frogs took a total of 99 hops, how many hops did it take the second frog to cross the road?
Answer: Let x represent the numb... | If Seattle has 20 sheep, Charleston has 4 * 20 sheep = <<20*4=80>>80 sheep
Toulouse has twice as many sheep as Charleston, which is 2 * 80 sheep = <<2*80=160>>160 sheep
Together, the three has 20 sheep + 160 sheep + 80 sheep = <<20+160+80=260>>260 sheep
#### 260 | {"answer": "If Seattle has 20 sheep, Charleston has 4 * 20 sheep = <<20*4=80>>80 sheep\nToulouse has twice as many sheep as Charleston, which is 2 * 80 sheep = <<2*80=160>>160 sheep\nTogether, the three has 20 sheep + 160 sheep + 80 sheep = <<20+160+80=260>>260 sheep\n#### 260", "question": "Toulouse has twice as many ... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:7 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 7
} | <|im_start|>user
Question: A building has four floors with ten rooms each. Legacy has to clean each room, and it takes her 6 hours to clean one room. If she earns $15 per hour of work, calculate the total amount of money she makes from cleaning all the floors in the building.
Answer: Since the building has four floors ... | First find how many gigabytes are in 40% of the file: 200 GB * 40% = <<200*40*.01=80>>80 GB
Then divide that number by the download rate to find the time until Windows restarts: 80 GB / 2 GB/minute = <<80/2=40>>40 minutes
Then find the time to download the whole file after the restart: 200 GB / 2 GB/minute = <<200/2=10... | {"answer": "First find how many gigabytes are in 40% of the file: 200 GB * 40% = <<200*40*.01=80>>80 GB\nThen divide that number by the download rate to find the time until Windows restarts: 80 GB / 2 GB/minute = <<80/2=40>>40 minutes\nThen find the time to download the whole file after the restart: 200 GB / 2 GB/minut... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:8 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 8
} | <|im_start|>user
Question: A classroom of 15 students turned in essays on Friday morning. The first 5 students each turned in essays with 2 pages. The next 5 students each turned in essays with 3 pages. The last 5 students each turned in essays with 1 page. What is the average page count per essay for the 15 students?
... | When he turned around he was 3*60=<<3*60=180>>180 miles from home
He was only able to drive 4-2=<<4-2=2>>2 hours in the first four hours
In half an hour he goes 30*.5=<<30*.5=15>>15 miles
He then drives another 2-.5=<<2-.5=1.5>>1.5 hours
In that time he goes 80*1.5=<<80*1.5=120>>120 miles
So he drove 120+15=<<120+15=13... | {"answer": "When he turned around he was 3*60=<<3*60=180>>180 miles from home\nHe was only able to drive 4-2=<<4-2=2>>2 hours in the first four hours\nIn half an hour he goes 30*.5=<<30*.5=15>>15 miles\nHe then drives another 2-.5=<<2-.5=1.5>>1.5 hours\nIn that time he goes 80*1.5=<<80*1.5=120>>120 miles\nSo he drove 1... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:9 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 9
} | <|im_start|>user
Question: Ali is a baker. Leila ordered 3 chocolate cakes for $12 each and 6 strawberry cakes for $22 each. How much should Leila pay Ali?
Answer: For the chocolate cakes: Leila should pay: 12 * 3 = $<<12*3=36>>36
For the strawberry cakes: Leila should pay: 22 * 6 = $<<22*6=132>>132
Combining the choc... | Eliza is entitled to 45 -40 = <<45-40=5>>5 hours overtime pay.
Her hourly rate for the overtime pay is $10 x 1.2 = $<<10*1.2=12>>12.
So, Eliza will receive $12 x 5 =$<<12*5=60>>60 for overtime pay.
Her regular weekly earning is $10 x 40 = $<<10*40=400>>400.
Thus, Eliza will receive a total of $400 + $60 = $<<400+60=460... | {"answer": "Eliza is entitled to 45 -40 = <<45-40=5>>5 hours overtime pay.\nHer hourly rate for the overtime pay is $10 x 1.2 = $<<10*1.2=12>>12.\nSo, Eliza will receive $12 x 5 =$<<12*5=60>>60 for overtime pay.\nHer regular weekly earning is $10 x 40 = $<<10*40=400>>400.\nThus, Eliza will receive a total of $400 + $60... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:10 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 10
} | <|im_start|>user
Question: One of the activities in Grade 6 PE class is to get the average weight of the students from each group. In one of the groups, the average weight of five girls is 45 kg while the average weight of five boys is 55 kg. What is the average weight of the ten students from that group?
Answer: The s... | The number of downloads of the program in the second month increased to 3*60 = <<3*60=180>>180
In the first two months, the total number of downloads of the program was 180+60 = <<180+60=240>>240
In the third month, the number of downloads of the program reduced by 30/100*180 = <<30/100*180=54>>54
There were 180-54 = <... | {"answer": "The number of downloads of the program in the second month increased to 3*60 = <<3*60=180>>180\nIn the first two months, the total number of downloads of the program was 180+60 = <<180+60=240>>240\nIn the third month, the number of downloads of the program reduced by 30/100*180 = <<30/100*180=54>>54\nThere ... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:11 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 11
} | <|im_start|>user
Question: Jerome has 20 classmates on his cell phone contact list. He has half as many out of school friends as classmates on his contact list. He also has his two parents and his sister on his contact list. How many people are on Jerome's contact list?
Answer: Jerome has 20 classmates on the list, ... | The total charge for the doughnuts was 3 x $68 = $<<3*68=204>>204.
The total charge for the mini cupcakes was 2 x $80 = $<<2*80=160>>160.
The total charge for the mini cheesecakes was 6 x $55 = $<<6*55=330>>330.
Therefore the total amount Toula paid for the pastries was $204 + $160 + $330 = $<<204+160+330=694>>694.
###... | {"answer": "The total charge for the doughnuts was 3 x $68 = $<<3*68=204>>204.\nThe total charge for the mini cupcakes was 2 x $80 = $<<2*80=160>>160.\nThe total charge for the mini cheesecakes was 6 x $55 = $<<6*55=330>>330.\nTherefore the total amount Toula paid for the pastries was $204 + $160 + $330 = $<<204+160+33... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:12 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 12
} | <|im_start|>user
Question: On Friday, Remy paid Sally’s Woodworking LLC a total of $20,700 for 150 pieces of furniture. Later that evening, the company’s accountant discovered that a new intern in the sales department had overcharged Remy. If the cost of a piece of furniture is $134, how much money will Sally’s Woodwor... | He makes $10.5 selling lemons each year because 7 x 1.5 = <<7*1.5=10.5>>10.5
He earns $7.5 each year from the lemon tree because 10.5 - 3 = <<10.5-3=7.5>>7.5
It will take 12 years to earn enough to pay off the tree because 90 / 7.5 = <<90/7.5=12>>12
He will make money in year 13 because 12 + 1 = <<12+1=13>>13
#### 13 | {"answer": "He makes $10.5 selling lemons each year because 7 x 1.5 = <<7*1.5=10.5>>10.5\nHe earns $7.5 each year from the lemon tree because 10.5 - 3 = <<10.5-3=7.5>>7.5\nIt will take 12 years to earn enough to pay off the tree because 90 / 7.5 = <<90/7.5=12>>12\nHe will make money in year 13 because 12 + 1 = <<12+1=1... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:13 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 13
} | <|im_start|>user
Question: Henry needs to declutter his book collection of 99 books. From his bookshelf, he was able to fill 3 boxes of 15 books each to take to a donation center. He had 21 books in a room he could donate, 4 on his coffee table and 18 cookbooks stashed in the kitchen. When he dropped them off, he no... | First multiply the five remaining vacuum cleaners by two to find out how many Melanie had before she visited the orange house: 5 * 2 = <<5*2=10>>10
Then add two to figure out how many vacuum cleaners she had before visiting the red house: 10 + 2 = <<10+2=12>>12
Now we know that 2/3 * x = 12, where x is the number of va... | {"answer": "First multiply the five remaining vacuum cleaners by two to find out how many Melanie had before she visited the orange house: 5 * 2 = <<5*2=10>>10\nThen add two to figure out how many vacuum cleaners she had before visiting the red house: 10 + 2 = <<10+2=12>>12\nNow we know that 2/3 * x = 12, where x is th... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:14 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 14
} | <|im_start|>user
Question: Arven bought five puppies for a total cost of $800. Three puppies are on sale for $150 each. How much does each of those two other puppies cost if they cost the same?
Answer: Three puppies on sale cost $150 x 3 = $<<150*3=450>>450.
The other two puppies cost $800 - $450 = $<<800-450=350>>350.... | There are 20 x 20/100 = <<20*20/100=4>>4 students who enrolled in contemporary dance.
So, 20 - 4 = <<20-4=16>>16 students are enrolled in either jazz or hip-hop dance.
There are 16 x 25/100 = <<16*25/100=4>>4 students who enrolled in jazz dance.
Hence, 16 - 4 = <<16-4=12>>12 students enrolled in hip-hop dance.
This is ... | {"answer": "There are 20 x 20/100 = <<20*20/100=4>>4 students who enrolled in contemporary dance.\nSo, 20 - 4 = <<20-4=16>>16 students are enrolled in either jazz or hip-hop dance.\nThere are 16 x 25/100 = <<16*25/100=4>>4 students who enrolled in jazz dance.\nHence, 16 - 4 = <<16-4=12>>12 students enrolled in hip-hop ... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:15 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 15
} | <|im_start|>user
Question: Aliens and Martians have different numbers of limbs. Aliens have three arms and eight legs, while Martians have half as many legs and twice as many arms. How many more limbs will five aliens have than five Martians?
Answer: Martians have 2 * 3 = <<2*3=6>>6 arms each.
They have 8 / 2 = <<8/2=4... | If he purchases jewelry, he will make a profit of 2.5% which is $5000*(2.5/100) = $<<5000*(2.5/100)=125>>125
If he purchases electronic gadgets, he will make a profit of 1.2% which is $8000*(1.2/100) = $<<8000*(1.2/100)=96>>96
If he wants to maximize profit, since $125 > $96, he will choose to purchase jewelry, thereby... | {"answer": "If he purchases jewelry, he will make a profit of 2.5% which is $5000*(2.5/100) = $<<5000*(2.5/100)=125>>125\nIf he purchases electronic gadgets, he will make a profit of 1.2% which is $8000*(1.2/100) = $<<8000*(1.2/100)=96>>96\nIf he wants to maximize profit, since $125 > $96, he will choose to purchase je... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:16 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 16
} | <|im_start|>user
Question: Jack has $43 in his piggy bank. He also gets an allowance of $10 a week. If Jack puts half of his allowance into his piggy bank every week, after 8 weeks how much will Jack have in his piggy bank?
Answer: Jack saves $10 / 2 = $<<10/2=5.00>>5.00 a week.
Over 8 weeks Jack saves $5.00/week x 8 w... | On the first day, the trains covered 2 trains * 80 miles/train = <<2*80=160>>160 miles together.
They also covered 150 miles/train * 2 trains = <<150*2=300>>300 miles together on the second day.
The combined distance the two trains covered in the two days is 300 miles + 160 miles = <<300+160=460>>460 miles
The average ... | {"answer": "On the first day, the trains covered 2 trains * 80 miles/train = <<2*80=160>>160 miles together.\nThey also covered 150 miles/train * 2 trains = <<150*2=300>>300 miles together on the second day.\nThe combined distance the two trains covered in the two days is 300 miles + 160 miles = <<300+160=460>>460 mile... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:17 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 17
} | <|im_start|>user
Question: Mona bikes 30 miles each week to stay fit. This week, she biked on Monday, Wednesday, and Saturday. On Wednesday, she biked 12 miles. On Saturday, she biked twice as far as on Monday. How many miles did she bike on Monday?
Answer: Let M be the number of miles Mona biked on Monday.
Thus, Mona ... | First find the total amount Jill makes per week teaching: $20/hour * 35 hours/week = $<<20*35=700>>700/week
Then find the total amount Jill makes per week coaching: $30/hour * 15 hours/week = $<<30*15=450>>450/week
Then add those two amounts to find the total amount Jill makes per week: $700/week + $450/week = $<<700+4... | {"answer": "First find the total amount Jill makes per week teaching: $20/hour * 35 hours/week = $<<20*35=700>>700/week\nThen find the total amount Jill makes per week coaching: $30/hour * 15 hours/week = $<<30*15=450>>450/week\nThen add those two amounts to find the total amount Jill makes per week: $700/week + $450/w... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:18 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 18
} | <|im_start|>user
Question: 25% of oysters have pearls in them. Jamie can collect 16 oysters during each dive. How many dives does he have to make to collect 56 pearls?
Answer: First find how many pearls Jamie collects per dive by multiplying the number of oysters by the chance each oyster has a pearl: .25 pearls/oyster... | She eats 3 eggs every day and there are 7 days in a week so she eats 3*7 = <<3*7=21>>21 eggs a week
After 4 weeks she will have eaten 4*21 = <<4*21=84>>84 eggs
There are 12 eggs in 1 dozen and she'll eat 84 eggs so that's 84/12 = <<84/12=7>>7 dozen eggs
#### 7 | {"answer": "She eats 3 eggs every day and there are 7 days in a week so she eats 3*7 = <<3*7=21>>21 eggs a week\nAfter 4 weeks she will have eaten 4*21 = <<4*21=84>>84 eggs\nThere are 12 eggs in 1 dozen and she'll eat 84 eggs so that's 84/12 = <<84/12=7>>7 dozen eggs\n#### 7", "question": "Claire makes a 3 egg omelet e... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:19 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 19
} | <|im_start|>user
Question: Joe and Adam built a garden wall with three courses of bricks. They realized the wall was too low and added 2 more courses. If each course of the wall had 400 bricks, and they took out half of the bricks in the last course to allow easy checkup of the garden, calculate the total number of bri... | First figure out how many hours it takes to hike a 12-mile trail at 4 mph by dividing the distance by the speed: 12 miles / 4 mph = <<12/4=3>>3 hours
Next subtract the time Marissa already spent walking to find out how much time she has left: 3 hours - 1 hour - 1 hour = <<3-1-1=1>>1 hour
Now figure out how much distanc... | {"answer": "First figure out how many hours it takes to hike a 12-mile trail at 4 mph by dividing the distance by the speed: 12 miles / 4 mph = <<12/4=3>>3 hours\nNext subtract the time Marissa already spent walking to find out how much time she has left: 3 hours - 1 hour - 1 hour = <<3-1-1=1>>1 hour\nNow figure out ho... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:20 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 20
} | <|im_start|>user
Question: The Reptile House at the zoo has 5 fewer animals than 3 times the number of animals housed in the Rain Forest exhibit. If the Reptile House has 16 animals, how many are in the Rain Forest exhibit?
Answer: Let x be the number of animals in the Rain Forest exhibit
3x-5=16
3x=21
x=<<7=7>>7
#### ... | There are 15 x 3/5 = <<15*3/5=9>>9 liters of water from the 15 liters pineapple drink.
After 1 liter of orange drink was spilled, there were 10 - 1 = <<10-1=9>>9 liters of orange drink left.
Out of the 9 liters, 9 x 2/3 = <<9*2/3=6>>6 liters are water.
Thus, there are a total of 9 + 6 = <<9+6=15>>15 liters of water out... | {"answer": "There are 15 x 3/5 = <<15*3/5=9>>9 liters of water from the 15 liters pineapple drink.\nAfter 1 liter of orange drink was spilled, there were 10 - 1 = <<10-1=9>>9 liters of orange drink left.\nOut of the 9 liters, 9 x 2/3 = <<9*2/3=6>>6 liters are water.\nThus, there are a total of 9 + 6 = <<9+6=15>>15 lite... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:21 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 21
} | <|im_start|>user
Question: Francine wants to know if a local chemical plant is releasing toxic waste that makes the frogs mutate. In one local pond, she catches 5 frogs with extra legs, 2 frogs with 2 heads, two frogs that are bright red, and 18 normal frogs. What percentage of the frogs have mutated, rounded to the ne... | When Raymond's son was born Samantha was 23 - 6 = <<23-6=17>>17 years old.
Thus it has been 31 - 17 = <<31-17=14>>14 years since Raymond's son was born.
#### 14 | {"answer": "When Raymond's son was born Samantha was 23 - 6 = <<23-6=17>>17 years old.\nThus it has been 31 - 17 = <<31-17=14>>14 years since Raymond's son was born.\n#### 14", "question": "Raymond and Samantha are cousins. Raymond was born 6 years before Samantha. Raymond had a son at the age of 23. If Samantha is now... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:22 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 22
} | <|im_start|>user
Question: There is a fruit display that has apples, oranges, and bananas. There are twice as many apples as oranges, and there are twice as many oranges as bananas. How many fruits are on the display in total if there are 5 bananas on the display?
Answer: We start by finding out that since there are ... | His first 3 customers buy 3 * 1 = <<3*1=3>>3 DVDs.
His next 2 buy 2 * 2 = <<2*2=4>>4 DVDs.
He sells a total of 3 + 4 + 0 = <<3+4+0=7>>7 DVDs.
#### 7 | {"answer": "His first 3 customers buy 3 * 1 = <<3*1=3>>3 DVDs.\nHis next 2 buy 2 * 2 = <<2*2=4>>4 DVDs.\nHe sells a total of 3 + 4 + 0 = <<3+4+0=7>>7 DVDs.\n#### 7", "question": "Billy sells DVDs. He has 8 customers on Tuesday. His first 3 customers buy one DVD each. His next 2 customers buy 2 DVDs each. His last 3 c... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:23 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 23
} | <|im_start|>user
Question: Rick has 130 cards. He decided to only keep 15 cards and so he gave Miguel some of the cards. Then, he saw 8 friends and decided to give them 12 cards each, and the remaining cards were given equally to his 2 sisters. If each of Rick's sisters got 3 cards, how many cards did Rick give to Migu... | The candle burns for 5 - 1 = <<5-1=4>>4 hours.
Thus, the candle will be 2 * 4 = <<2*4=8>>8 centimeters shorter.
#### 8 | {"answer": "The candle burns for 5 - 1 = <<5-1=4>>4 hours.\nThus, the candle will be 2 * 4 = <<2*4=8>>8 centimeters shorter.\n#### 8", "question": "A candle melts by 2 centimeters every hour that it burns. How many centimeters shorter will a candle be after burning from 1:00 PM to 5:00 PM?"} | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:24 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 24
} | <|im_start|>user
Question: Hannah has three dogs. The first dog eats 1.5 cups of dog food a day. The second dog eats twice as much while the third dog eats 2.5 cups more than the second dog. How many cups of dog food should Hannah prepare in a day for her three dogs?
Answer: The second dog eats 1.5 x 2 = <<1.5*2=3>>3 c... | Let X be the original price of the book. The discounted price is X - X*25% = $19.50.
Combining like terms, we get .75X = $19.50
Dividing both sides by .75, we get X = $26.
#### 26 | {"answer": "Let X be the original price of the book. The discounted price is X - X*25% = $19.50.\nCombining like terms, we get .75X = $19.50\nDividing both sides by .75, we get X = $26.\n#### 26", "question": "Kyle bought last year's best-selling book for $19.50. This is with a 25% discount from the original price. Wha... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:25 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 25
} | <|im_start|>user
Question: A boy runs 1.5 km in 45 minutes. What is his speed in kilometers per hour?
Answer: Since 1 hour is equal to 60 minutes, then 45 minutes is equal to 45/60 = <<45/60=0.75>>0.75 hours.
Then, the speed in kilometers per hour is 1.5 km/0.75 hour = 2 kilometers per hour.
#### 2
Question: Tall Tuna... | Five packs of milk cost $3 x 5 = $<<3*5=15>>15.
Four apples cost $1.50 x 4 = $<<1.5*4=6>>6.
The total cost of the chicken meal, milk, and apples is $12 + $15 + $6 = $<<12+15+6=33>>33.
Thus, the boxes of pizza cost $50 - $33 = $<<50-33=17>>17.
Therefore, Marie ordered $17/$8.50 = <<17/8.5=2>>2 boxes of pizza.
#### 2 | {"answer": "Five packs of milk cost $3 x 5 = $<<3*5=15>>15.\nFour apples cost $1.50 x 4 = $<<1.5*4=6>>6.\nThe total cost of the chicken meal, milk, and apples is $12 + $15 + $6 = $<<12+15+6=33>>33.\nThus, the boxes of pizza cost $50 - $33 = $<<50-33=17>>17.\nTherefore, Marie ordered $17/$8.50 = <<17/8.5=2>>2 boxes of p... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:26 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 26
} | <|im_start|>user
Question: Marlon had 42 lollipops in the beginning. On his way home, he saw Emily and gave her 2/3 of his lollipops. Then, Marlon kept 4 lollipops and gave the rest to Lou. How many lollipops did Lou receive?
Answer: Marlon gave Emily 42 x 2/3 = <<42*2/3=28>>28 lollipops.
So, he only had 42 - 28 = <<42... | 3 * (16.50 + 22.50 + 42) = $<<3*(16.50+22.50+42)=243>>243
Mishka spent $243 on clothing.
#### 243 | {"answer": "3 * (16.50 + 22.50 + 42) = $<<3*(16.50+22.50+42)=243>>243\nMishka spent $243 on clothing.\n#### 243", "question": "Mishka bought 3 pairs of shorts, 3 pairs of pants, and 3 pairs of shoes. One pair of shorts costs $16.50. One pair of pants costs $22.50 and one pair of shoes costs $42. How many dollars did Mi... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:27 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 27
} | <|im_start|>user
Question: Marlene wants to buy half a dozen of shirts to avail of the sale. The regular price of a shirt is $50 and is now on sale at a 20% discount. How much will Marlene pay for the shirts?
Answer: Marlene wants to buy 12/2 = <<12/2=6>>6 shirts.
The discount price of the shirt is $50 x 20/100 = $<<50... | Each container of ice cream has 15 servings and she eats 1 serving a night so after 60 days she will need 60/15 = 4 containers of ice cream
If each carton costs $4.00 and she needs 4 containers then it will cost her 4*4 = $<<4*4=16.00>>16.00
#### 16 | {"answer": "Each container of ice cream has 15 servings and she eats 1 serving a night so after 60 days she will need 60/15 = 4 containers of ice cream\nIf each carton costs $4.00 and she needs 4 containers then it will cost her 4*4 = $<<4*4=16.00>>16.00\n#### 16", "question": "Cynthia eats one serving of ice cream eve... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:28 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 28
} | <|im_start|>user
Question: There are 50 passengers on a bus. At the first stop, 16 more passengers get on the bus. On the other stops, 22 passengers get off the bus and 5 passengers more get on the bus. How many passengers are there on the bus in total at the last station?
Answer: There are 50+16=<<50+16=66>>66 passen... | He traveled 20 miles + 15 miles = <<20+15=35>>35 miles not counting the distance between stops.
Henry traveled 60 miles - 35 miles = <<60-35=25>>25 miles between his first and second stop.
#### 25 | {"answer": "He traveled 20 miles + 15 miles = <<20+15=35>>35 miles not counting the distance between stops.\nHenry traveled 60 miles - 35 miles = <<60-35=25>>25 miles between his first and second stop.\n#### 25", "question": "Henry made two stops during his 60-mile bike trip. He first stopped after 20 miles. His second... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:29 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 29
} | <|im_start|>user
Question: Marlon has a gift card for $200. He spent half of it on Monday and one-fourth of the remainder on Tuesday. How much was left on his gift card?
Answer: Since Marlon spent half of the gift card, then half of it which is $200/2 = $<<200/2=100>>100 remained on Monday.
Then on Tuesday, he spent $1... | The second pair of heels costs 33 * 2 = $<<33*2=66>>66.
The heels together cost 66 + 33 = $<<66+33=99>>99.
The boots cost $5 more than both pairs of heels together, so the boots cost 99 + 5 = $104.
#### 104 | {"answer": "The second pair of heels costs 33 * 2 = $<<33*2=66>>66.\nThe heels together cost 66 + 33 = $<<66+33=99>>99.\nThe boots cost $5 more than both pairs of heels together, so the boots cost 99 + 5 = $104.\n#### 104", "question": "Gloria is shoe shopping when she comes across a pair of boots that fit her shoe bud... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:30 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 30
} | <|im_start|>user
Question: Ben's new car cost twice as much as his old car. He sold his old car for $1800 and used that to pay off some of the cost of his new car. He still owes another $2000 on his new car. How much did his old car cost, in dollars?
Answer: His new car costs 1800+2000=<<1800+2000=3800>>3800 dollars.
H... | The total ratio representing their ages is 7+11= <<7+11=18>>18
Since the fraction of the ratio that represents Allen's age is 11/18, Allen's current age is 11/18*162 = <<11/18*162=99>>99
If Allen is currently 99 years old, in 10 years he will be 99+10 = <<99+10=109>>109 years old
#### 109 | {"answer": "The total ratio representing their ages is 7+11= <<7+11=18>>18\nSince the fraction of the ratio that represents Allen's age is 11/18, Allen's current age is 11/18*162 = <<11/18*162=99>>99\nIf Allen is currently 99 years old, in 10 years he will be 99+10 = <<99+10=109>>109 years old\n#### 109", "question": "... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:31 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 31
} | <|im_start|>user
Question: Peyton Manning throws a football 20 yards when the temperature is 50 degrees Fahrenheit, but the ball moves twice as far when the temperature is 80 degrees Fahrenheit. Last Saturday, the temperature was 50 degrees Fahrenheit, and he threw the ball twenty times. On Sunday, the temperature rose... | The second person guessed 60 because 20 + (80/2)= <<20+(80/2)=60>>60
The second guessed 100 because 80 x 1.25 = <<80*1.25=100>>100
The total amount guessed is 240 because 80 + 60 + 100 = <<80+60+100=240>>240
The average amount guessed is 80 because 240 / 3 = <<240/3=80>>80
#### 80 | {"answer": "The second person guessed 60 because 20 + (80/2)= <<20+(80/2)=60>>60\nThe second guessed 100 because 80 x 1.25 = <<80*1.25=100>>100\nThe total amount guessed is 240 because 80 + 60 + 100 = <<80+60+100=240>>240\nThe average amount guessed is 80 because 240 / 3 = <<240/3=80>>80\n#### 80", "question": "Gunter ... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:32 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 32
} | <|im_start|>user
Question: Conor can chop 12 eggplants, 9 carrots, and 8 potatoes in a day. If he works 4 times a week, how many vegetables can he chop?
Answer: Conor can chop 12 x 4 = <<12*4=48>>48 eggplants a week.
He can chop 9 x 4 = <<9*4=36>>36 carrots a week.
He can chop 8 x 4 = <<8*4=32>>32 potatoes a week.
Ther... | He spends 10*.5=<<10*.5=5>>5 hours per day
That means he spends 5*7=<<5*7=35>>35 hours per week
#### 35 | {"answer": "He spends 10*.5=<<10*.5=5>>5 hours per day\nThat means he spends 5*7=<<5*7=35>>35 hours per week\n#### 35", "question": "John takes care of 10 dogs. Each dog takes .5 hours a day to walk and take care of their business. How many hours a week does he spend taking care of dogs?"} | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:33 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 33
} | <|im_start|>user
Question: Jason has a carriage house that he rents out. He’s charging $50.00 per day or $500.00 for 14 days. Eric wants to rent the house for 20 days. How much will it cost him?
Answer: He wants to rent for 20 days and there is a deal if you rent for 14 days so that leaves 20-14 = <<20-14=6>>6 indiv... | Let x be the number of silver coins Gretchen has
Gretchen has x+30 gold coins.
x+x+30=110
2*x=80
x=<<40=40>>40
Gretchen has 40+30=<<40+30=70>>70 gold coins
#### 70 | {"answer": "Let x be the number of silver coins Gretchen has\nGretchen has x+30 gold coins.\nx+x+30=110\n2*x=80\nx=<<40=40>>40\nGretchen has 40+30=<<40+30=70>>70 gold coins\n#### 70", "question": "Gretchen has 110 coins. There are 30 more gold coins than silver coins. How many gold coins does Gretchen have?"} | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:34 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 34
} | <|im_start|>user
Question: A quantity surveyor is figuring the construction costs for a couple that wishes to build a house. The costs are as follows: land costs $50 per square meter, bricks cost $100 per 1000 bricks and roof tiles cost $10 per roof tile. If the house they wish to build requires 2000 square meters, 100... | Half of Raymond's jewels is 40/2 = <<40/2=20>>20
Since Aaron has 5 more jewels than half of Raymond's jewels, he has 20+5 = <<5+20=25>>25 jewels.
If Siobhan has 2 fewer jewels than Aaron, he has 25-2 = 23 jewels.
#### 23 | {"answer": "Half of Raymond's jewels is 40/2 = <<40/2=20>>20\nSince Aaron has 5 more jewels than half of Raymond's jewels, he has 20+5 = <<5+20=25>>25 jewels.\nIf Siobhan has 2 fewer jewels than Aaron, he has 25-2 = 23 jewels.\n#### 23", "question": "Siobhan has 2 fewer jewels than Aaron. Aaron has 5 more jewels than h... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:35 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 35
} | <|im_start|>user
Question: If John travels 15 miles on a bike ride, and Jill travels 5 miles less, how many miles does Jim travel if he travels only 20% as far as Jill?
Answer: To find how far Jim travels, we must first find out how far Jill did. To do this, we perform 15-5=<<15-5=10>>10 miles in total for Jill.
We the... | He scored 4*.25=<<4*.25=1>>1 point more in the second 20 minutes
So he scored 4+1=<<4+1=5>>5 points in the second 20 minutes
So in total, he scored 4+5=<<4+5=9>>9 points in the 40 minutes
#### 9 | {"answer": "He scored 4*.25=<<4*.25=1>>1 point more in the second 20 minutes\nSo he scored 4+1=<<4+1=5>>5 points in the second 20 minutes\nSo in total, he scored 4+5=<<4+5=9>>9 points in the 40 minutes\n#### 9", "question": "Mike plays ping pong for 40 minutes. In the first 20 minutes, he scores 4 points. In the seco... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:36 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 36
} | <|im_start|>user
Question: The GooGoo brand of clothing manufactures two types of shirts, one with 3 buttons and the other with 5 buttons. A large department store ordered GooGoo to manufacture 200 of each type of shirt. How many buttons will the GooGoo brand use to manufacturer all of the shirts for this order?
Answ... | He eats 2 yogurts a day so over 30 days he will need 2*30 = <<2*30=60>>60 yogurts
The yogurt is sold as 4 units for $5.00 and he needs 60 yogurts so that’s 60/4 = <<60/4=15>>15 4-packs
Each 4-pack is $5.00 and he needs 15 4-packs so he will spend $5*15 = $<<5*15=75.00>>75.00 on yogurt
#### 75 | {"answer": "He eats 2 yogurts a day so over 30 days he will need 2*30 = <<2*30=60>>60 yogurts\nThe yogurt is sold as 4 units for $5.00 and he needs 60 yogurts so that’s 60/4 = <<60/4=15>>15 4-packs\nEach 4-pack is $5.00 and he needs 15 4-packs so he will spend $5*15 = $<<5*15=75.00>>75.00 on yogurt\n#### 75", "question... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:37 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 37
} | <|im_start|>user
Question: There are 24 students in the class. One-third had their school portraits taken before lunch. After lunch, but before gym class, 10 additional students had their portraits taken. After gym class, how many students have not yet had their picture taken?
Answer: Before lunch, 24/3=<<24/3=8>>8 stu... | He spent $160 on video games because 8 x 20 = <<8*20=160>>160
He earned $165 from his sale because 160 + 5 = <<160+5=165>>165
He sold 11 lego sets because 11 x 15 = <<11*15=165>>165
He has 2 sets left because 13 - 11 = <<13-11=2>>2
#### 2 | {"answer": "He spent $160 on video games because 8 x 20 = <<8*20=160>>160\nHe earned $165 from his sale because 160 + 5 = <<160+5=165>>165\nHe sold 11 lego sets because 11 x 15 = <<11*15=165>>165\nHe has 2 sets left because 13 - 11 = <<13-11=2>>2\n#### 2", "question": "John plans to sell all his toys and use the money ... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... | |
gsm8k:38 | GSM8K | gsm8k | {
"dataset": "openai/gsm8k",
"config": "main",
"split": "test",
"doc_id": 38
} | <|im_start|>user
Question: The town’s annual budget totals $32 million. If half of the budget goes towards policing and $12 million goes towards education. How much money is left for managing public spaces?
Answer: The annual budget for policing is 32 / 2 = $<<32/2=16>>16 million.
The combined budget for education and ... | He has 7-1=<<7-1=6>>6 possible running days.
He runs half of those so he runs 6/2=<<6/2=3>>3 days.
So for two days, he runs 3/2=<<3/2=1.5>>1.5 hours.
So in total, he runs 3+1.5+1.5=<<3+1.5+1.5=6>>6 hours.
So he runs 60/6=<<60/6=10>>10 mph.
#### 10 | {"answer": "He has 7-1=<<7-1=6>>6 possible running days.\nHe runs half of those so he runs 6/2=<<6/2=3>>3 days.\nSo for two days, he runs 3/2=<<3/2=1.5>>1.5 hours.\nSo in total, he runs 3+1.5+1.5=<<3+1.5+1.5=6>>6 hours.\nSo he runs 60/6=<<60/6=10>>10 mph.\n#### 10", "question": "John runs 60 miles a week. He runs 3 day... | Qwen3 chat template; enable_thinking=True; add_generation_prompt=True | 5 | {
"do_sample": true,
"temperature": 0.6000000238418579,
"top_p": 0.949999988079071,
"top_k": 20,
"min_p": 0,
"max_gen_toks": 16384,
"seed": 42,
"stop_sequences": [
"Question:",
"</s>",
"<|im_end|>"
]
} | {
"metric": "exact_match",
"metric_implementation": "lm_eval.api.metrics.exact_match_hf_evaluate",
"thinking_end_token": "</think>",
"thinking_extraction_regex": "(?s)^.*</think>\\s*",
"thinking_extraction_rule": "Take response.split('</think>')[-1].lstrip(); if the marker is absent, keep the complete respons... |
End of preview. Expand in Data Studio
Latent Reasoning Evaluation 2K
This dataset is the exact 2,135-request slice used by the project’s fixed-compression Qwen3 evaluation. It contains at most 500 examples per benchmark: GSM8K (500), MATH-500 (500), BBH causal judgement (187), BBH sports understanding (250), GPQA diamond (198), and MMLU-Pro physics (500).
Every prompt is already fully rendered: benchmark formatting, deterministic few-shot examples where applicable, and the Qwen3 chat template (enable_thinking=True, add_generation_prompt=True) are included. Send prompt directly to a Qwen3-compatible model; do not add another chat template or few-shot examples.
Loading
from datasets import load_dataset
ds = load_dataset("pihull/latent-reasoning-evaluation-2k", split="train")
example = ds[0]
print(example["prompt"])
print(example["reference_answer"])
Columns
id,benchmark,task_name: stable evaluation identifiers.source: original HF dataset/config/split and source document index.prompt: exact Qwen3-formatted model input used in evaluation.reference_answer: the task’s gold target.reference_solution: MATH-500 worked solution; required for its semantic scorer and empty for other tasks.source_record_json: serialized processed source record, including options and benchmark metadata where available.generation: the shared sampled decoding parameters and per-task stop sequences.assessment: a structured scorer specification. It records the post-thinking extraction rule, stops, exact-match normalizations, and the gold field to use.
Scoring a model response
The runner first removes reasoning, then applies task-specific stops:
def processed_response(raw_response: str, assessment: dict) -> str:
# Exact lm-eval behavior: when the marker is absent this retains raw_response.
text = raw_response.split(assessment["thinking_end_token"])[-1].lstrip()
for stop in assessment["stop_sequences"]:
if stop:
text = text.split(stop)[0]
return text
assessment["thinking_extraction_regex"] ((?s)^.*</think>\\s*) is the equivalent regex form when </think> is present. For GSM8K/BBH/GPQA/MMLU-Pro, score the processed response with lm_eval.api.metrics.exact_match_hf_evaluate, using assessment["metric_kwargs_json"]; this preserves each benchmark’s case, punctuation, and regex normalization rules. In particular, GSM8K uses (?s).*#### to discard a chain of thought before its final #### answer.
For MATH-500, use the exact semantic scorer:
from math_verify import parse, verify
correct = bool(verify(
gold=parse(example["reference_solution"]),
target=parse(processed_response(model_response, example["assessment"])),
))
The task source datasets retain their own licences and terms. This dataset republishes the selected evaluation material for reproducibility; consult and comply with the original dataset terms cited by each row’s source.dataset.
- Downloads last month
- 13