Dataset Viewer
Auto-converted to Parquet Duplicate
prompt
list
question
string
answer
string
metadata
string
model_responses
sequence
model_responses__initial_answer
sequence
prompt__initial_answer__metadata
string
model_responses__initial_answer__metadata
string
verification_prompt
list
model_responses__verification
sequence
verification_prompt__verification__metadata
string
model_responses__verification__metadata
string
task_source
string
model_responses__initial_answer__eval_is_correct
sequence
model_responses__initial_answer__eval_extracted_answers
sequence
model_responses__initial_answer__eval_extraction_metadata
string
model_responses__initial_answer__eval_evaluation_metadata
string
answer_is_correct
int64
model_thinks_answer_is_correct
int64
verification_score
float64
filtered_verification_score
float64
[ { "content": "Solve the problem step by step. Think step by step first.\nThe answer is a series of arithmetic operations (+, -, *, /) that results in the target number.\n\nWrite the final answer in <answer> </answer> tags.\nFor the final answer, make sure that each step in the final answer is written as Step X:...
Find a way to make 125 using all of these numbers: 96, 97, 68. Each number can only be used once. Final answer format instructions: 1. Provide your solution as a arithmetic expression (no '=' sign). 2. Do not include the target number in the expression. 3. Use '*' for multiplication. 4. Use '/' for division. 5. Do not...
97 + 96 - 68
"{\"source_dataset\": \"countdown\", \"source_index\": 5000, \"numbers\": [96, 97, 68], \"target\": 125, \"expression\": \"97 + 96 - 68\"}"
[ "To solve this problem, we need to combine the given numbers [96, 97, 68] using basic arithmetic operations (+, -, *, /) such that the result is 125. Each number should be used exactly once.\n\nHere’s how we can approach it:\n\n1. First, let's try adding some numbers together to get close to 125.\n - We know that...
[ "Step 1: 97 - 96 = 1\nStep 2: 1 * 68 = 68\nStep 3: 68 + 1 = 69\n\nTherefore, the final answer is:\n<answer>\nStep 1: 97 - 96 = 1\nStep 2: 1 * 68 = 68\nStep 3: 68 + 1 = 69\n</answer>" ]
{"api_url": "http://127.0.0.1:63496", "backend": "litellm", "chat_template_applied": true, "generation_params": {"chat_template_applied": true, "max_tokens": 768, "temperature": 0.7, "top_p": 0.8}, "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "prompt": [{"content": "Solve the problem step by step. Think step...
{"backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 1}
[ { "content": "Solve the problem step by step. Think step by step first.\nThe answer is a series of arithmetic operations (+, -, *, /) that results in the target number.\n\nWrite the final answer in <answer> </answer> tags.\nFor the final answer, make sure that each step in the final answer is written as Step X:...
[ "Yes" ]
{"api_url": "http://127.0.0.1:63496", "backend": "litellm", "chat_template_applied": true, "generation_params": {"chat_template_applied": true, "frequency_penalty": 2.0, "logit_bias": {"2753": 10, "9454": 10}, "max_tokens": 20, "temperature": 0.0, "top_p": 1.0}, "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "...
{"backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 1}
countdown
[ false ]
[ "Step 1: 97 - 96 = 1\nStep 2: 1 * 68 = 68\nStep 3: 68 + 1 = 69" ]
[{}]
[{"answer_block": "Step 1: 97 - 96 = 1\nStep 2: 1 * 68 = 68\nStep 3: 68 + 1 = 69", "error": "Line uses unavailable numbers: Step 3: 68 + 1 = 69", "final_answer": 69.0, "is_correct": false}]
1
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Calculate 105 using all of these numbers: 21, 80, 46.\nEach number may be used at most once.\n\nFin(...TRUNCATED)
46 + 80 - 21
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5001, \\\"numbers\\\": [21, 80,(...TRUNCATED)
["To solve this problem, we need to combine the given numbers using basic arithmetic operations to g(...TRUNCATED)
["To solve this problem, we need to use the numbers 21, 80, and 46 exactly once, and we can use the (...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ true ]
[ "Step 1: 46 + 80 = 126\nStep 2: 126 - 21 = 105" ]
[{}]
"[{\"answer_block\": \"Step 1: 46 + 80 = 126\\nStep 2: 126 - 21 = 105\", \"error\": null, \"final_an(...TRUNCATED)
2
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Using all the numbers 6, 76, 96, create an expression that equals 500.\nYou can only use each numbe(...TRUNCATED)
6*96 - 76
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5002, \\\"numbers\\\": [6, 76, (...TRUNCATED)
["To solve this problem using the numbers [6, 76, 96] and creating an equation that equals 500, we n(...TRUNCATED)
["Step 1: 6 * 76 = 456\nStep 2: 456 + 96 = 552\n\nTherefore, the final answer is:\n\n<answer>\nStep (...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ false ]
[ "Step 1: 6 * 76 = 456\nStep 2: 456 + 96 = 552" ]
[{}]
"[{\"answer_block\": \"Step 1: 6 * 76 = 456\\nStep 2: 456 + 96 = 552\", \"error\": \"Incorrect final(...TRUNCATED)
1
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Find a way to make 630 using all of these numbers: 42, 12, 3.\nEach number can only be used once.\n(...TRUNCATED)
42*(12 + 3)
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5003, \\\"numbers\\\": [42, 12,(...TRUNCATED)
["To solve this problem, we need to use the numbers [42, 12, 3] and apply basic arithmetic operation(...TRUNCATED)
["To solve this problem, we need to use the numbers [42, 12, 3] and create an equation that equals 6(...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ false ]
[ "Step 1: 42 / 3 = 14\nStep 2: 14 * 12 = 168\nStep 3: 168 + 14 = 182" ]
[{}]
"[{\"answer_block\": \"Step 1: 42 / 3 = 14\\nStep 2: 14 * 12 = 168\\nStep 3: 168 + 14 = 182\", \"err(...TRUNCATED)
1
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Find a way to make 194 using all of these numbers: 18, 85, 91.\nEach number can only be used once.\(...TRUNCATED)
85 + 91 + 18
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5004, \\\"numbers\\\": [18, 85,(...TRUNCATED)
["To solve the problem, we need to use the numbers [18, 85, 91] and the arithmetic operations (+, -,(...TRUNCATED)
["To solve this problem, we need to use the numbers [18, 85, 91] and create an equation that equals (...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ false ]
[ "Step 1: 91 * 2 = 182\nStep 2: 182 + 18 = 194" ]
[{}]
"[{\"answer_block\": \"Step 1: 91 * 2 = 182\\nStep 2: 182 + 18 = 194\", \"error\": \"Line uses unava(...TRUNCATED)
1
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Calculate 103 using all of these numbers: 86, 69, 52.\nEach number may be used at most once.\n\nFin(...TRUNCATED)
86 + 69 - 52
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5005, \\\"numbers\\\": [86, 69,(...TRUNCATED)
["To solve this problem, we need to use the numbers [86, 69, 52] and the operations (+, -, *, /) to (...TRUNCATED)
["Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103\n\n<answer>\nStep 1: 86 + 69 = 155\nStep 2: 155 - 52(...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ true ]
[ "Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103" ]
[{}]
"[{\"answer_block\": \"Step 1: 86 + 69 = 155\\nStep 2: 155 - 52 = 103\", \"error\": null, \"final_an(...TRUNCATED)
2
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Using all the numbers 63, 7, 87, create an expression that equals 354.\nYou can only use each numbe(...TRUNCATED)
63*7 - 87
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5006, \\\"numbers\\\": [63, 7, (...TRUNCATED)
["To solve the problem using the numbers [63, 7, 87], we need to find an equation that equals 354. L(...TRUNCATED)
["Step 1: 63 * 7 = 441\nStep 2: 441 - 87 = 354\n\n<answer>\nStep 1: 63 * 7 = 441\nStep 2: 441 - 87 =(...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ true ]
[ "Step 1: 63 * 7 = 441\nStep 2: 441 - 87 = 354" ]
[{}]
"[{\"answer_block\": \"Step 1: 63 * 7 = 441\\nStep 2: 441 - 87 = 354\", \"error\": null, \"final_ans(...TRUNCATED)
2
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Find a way to make 102 using all of these numbers: 8, 50, 19.\nEach number can only be used once.\n(...TRUNCATED)
19*8 - 50
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5007, \\\"numbers\\\": [8, 50, (...TRUNCATED)
["To solve the problem of creating an equation using the numbers [8, 50, 19] that equals 102, we can(...TRUNCATED)
["Step 1: 8 * 50 = 400\nStep 2: 400 - 19 = 381\nStep 3: 381 - 102 = 279\n\nThe final answer is: <ans(...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ false ]
[ "Step 1: 8 * 50 = 400\nStep 2: 400 - 19 = 381\nStep 3: 381 - 102 = 279" ]
[{}]
"[{\"answer_block\": \"Step 1: 8 * 50 = 400\\nStep 2: 400 - 19 = 381\\nStep 3: 381 - 102 = 279\", \"(...TRUNCATED)
1
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Calculate 124 using all of these numbers: 38, 46, 40.\nEach number may be used at most once.\n\nFin(...TRUNCATED)
40 + 46 + 38
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5008, \\\"numbers\\\": [38, 46,(...TRUNCATED)
["To solve this problem, we need to use the numbers [38, 46, 40] and the operations (+, -, *, /) to (...TRUNCATED)
["To solve this problem, we need to use the numbers [38, 46, 40] exactly once and combine them using(...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ true ]
[ "Step 1: 38 + 40 = 78\nStep 2: 78 + 46 = 124" ]
[{}]
"[{\"answer_block\": \"Step 1: 38 + 40 = 78\\nStep 2: 78 + 46 = 124\", \"error\": null, \"final_answ(...TRUNCATED)
2
2
0.232
0.232
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
"Calculate 132 using all of these numbers: 92, 21, 19.\nEach number may be used at most once.\n\nFin(...TRUNCATED)
19 + 92 + 21
"\"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5009, \\\"numbers\\\": [92, 21,(...TRUNCATED)
["To solve the problem of creating an equation using the numbers [92, 21, 19] that equals 132, we ca(...TRUNCATED)
["Step 1: 92 - 21 = 71\nStep 2: 71 - 19 = 52\nStep 3: 52 * 2 = 104\nStep 4: 104 + 28 = 132\n\nTheref(...TRUNCATED)
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
[{"content":"Solve the problem step by step. Think step by step first.\nThe answer is a series of ar(...TRUNCATED)
[ "Yes" ]
"{\"api_url\": \"http://127.0.0.1:63496\", \"backend\": \"litellm\", \"chat_template_applied\": true(...TRUNCATED)
"{\"backend\": \"litellm\", \"model_name\": \"hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct\", \"n_response(...TRUNCATED)
countdown
[ false ]
[ "Step 1: 92 - 21 = 71\nStep 2: 71 - 19 = 52\nStep 3: 52 * 2 = 104\nStep 4: 104 + 28 = 132" ]
[{}]
"[{\"answer_block\": \"Step 1: 92 - 21 = 71\\nStep 2: 71 - 19 = 52\\nStep 3: 52 * 2 = 104\\nStep 4: (...TRUNCATED)
1
2
0.232
0.232
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
4