Dataset Viewer
Auto-converted to Parquet Duplicate
question
stringlengths
7
989
answer
stringlengths
0
28
task_config
stringclasses
14 values
task_source
stringclasses
6 values
prompt
listlengths
1
1
model_responses
listlengths
0
0
model_responses__eval_is_correct
listlengths
0
0
all_other_columns
stringlengths
90
1.36k
original_split
stringclasses
1 value
acronym
stringclasses
341 values
answer_index
int64
0
4
answer_key
stringclasses
5 values
choices
dict
difficulty
stringclasses
1 value
domain
stringclasses
1 value
evaluation_type
stringclasses
1 value
expected_answer_format
stringclasses
1 value
formed_acronym
stringclasses
341 values
id
stringlengths
7
34
length
int64
4
8
letters
stringclasses
600 values
metadata
stringlengths
41
152
original_answer
stringlengths
48
1.07k
source
stringclasses
601 values
task_type
stringclasses
1 value
variant
stringclasses
1 value
word_count
int64
4
8
words
listlengths
4
8
model_responses__best_of_n_atags
listlengths
4
4
model_responses__best_of_n_atags__finish_reason_length_flags
listlengths
4
4
model_responses__best_of_n_atags__length_partial_responses
listlengths
4
4
prompt__best_of_n_atags__metadata
dict
model_responses__best_of_n_atags__metadata
dict
model_responses__best_of_n_atags__eval_is_correct
listlengths
4
4
model_responses__best_of_n_atags__eval_extracted_answers
listlengths
4
4
model_responses__best_of_n_atags__eval_extraction_metadata
stringlengths
463
5.09k
model_responses__best_of_n_atags__eval_evaluation_metadata
stringlengths
224
5.17k
model_responses__best_of_n_atags__internal_answers__eval_is_correct
listlengths
4
4
model_responses__best_of_n_atags__internal_answers__eval_extracted_answers
listlengths
4
4
model_responses__best_of_n_atags__internal_answers__eval_extraction_metadata
stringlengths
559
201k
model_responses__best_of_n_atags__internal_answers__eval_evaluation_metadata
stringlengths
229
167k
model_responses__best_of_n_atags__metrics
dict
eval_date
stringdate
2025-11-18 00:00:00
2025-11-18 00:00:00
split
stringclasses
1 value
revision_name
stringclasses
1 value
model_path
stringclasses
1 value
checkpoint_step
int64
0
0
stage_name
stringclasses
1 value
stage_number
int64
1
1
timestamp
stringdate
2025-11-18 22:20:05
2025-11-18 22:20:05
eval_repo_id
stringclasses
1 value
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [96, 97, 68], create an equation that equals 125. You can use basic arithmetic operations (+, -, *, /) and each number can o...
97 + 96 - 68
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [96, 97, 68], create an equation that equals 125. You can use basic arithmetic operations (+, -, *,...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5000, \"numbers\": [96, 97, 68], \"target\": 125, \"expression\": \"97 + 96 - 68\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5000, "numbers": [96, 97, 68], "target": 125, "expression": "97 + 96 - 68"}
null
null
null
null
null
null
[ "To solve this problem using the numbers from the list [96, 97, 68] such that their combination through basic arithmetic operations yields 125:\n\n**Step-by-step approach:**\n\nFirstly, let’s consider how we might combine these numbers effectively.\n- We notice that none of the direct additions or subtractions will...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, true, false, false ]
[ "97 + (68 - 96)", "97 + 96 - 68 = 125", "Step 1: Start with 96.\nStep 2: Add 97 to 96 to get 193.\nStep 3: Subtract 68 from 193 to get 125.", "Step 1: Start with the largest number, 97.\nStep 2: Add 96 to 97 to get 193.\nStep 3: Subtract 68 from 193 to get 125." ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 1741, "span_end": 1757, "original_span": "97 + (68 - 96)"}, {"extraction_type": "final", "total_spans_found": 6, "non_overlapping_spans": 6, "span_start": 8941, "span_end": 8961, "original_span": "97 + 96 - 68 = 125"}, {"ext...
[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "97 + (68 - 96)", "final_answer": 125.0}, {"is_correct": true, "error": null, "answer_block": "97 + 96 - 68 = 125", "final_answer": 125.0}, {"is_correct": false, "error": "Error evaluating final answer", "answer_block": "Step 1: ...
[ [ false ], [ false, true, true, true, true, true ], [ false ], [ false ] ]
[ [ "97 + (68 - 96)" ], [ "Step 1: 97 + 68 = 165\nStep 2: 165 - 40 = 125\n\nTherefore, the equation is:\n97 + 68 - 40 = 125", "96 + 97 - 68 = 125", "96 + (97 - 68) = 125", "97 + 96 - 68 = 125", "Step 1: 96 + 97 = 193\nStep 2: 193 - 68 = 125", "97 + 96 - 68 = 125" ], [ "Step 1: ...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1741, "span_text": "97 + (68 - 96)", "extraction_type": "internal", "span_start": 1741, "span_end": 1757, "original_span": "97 + (68 - 96)"}], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 1822, "span_text": "Step 1: 97 + 68 = 165\nStep 2:...
[[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "97 + (68 - 96)", "final_answer": 125.0}], [{"is_correct": false, "error": "Line uses unavailable numbers: Step 2: 165 - 40 = 125", "answer_block": "Step 1: 97 + 68 = 165\nStep 2: 165 - 40 = 125\n\nTherefore, the equation is:\n9...
{ "flips_by": [ 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 1, "pass_at_n": 1, "percent_correct": 0.25, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, ...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776829
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [21, 80, 46], create an equation that equals 105. You can use basic arithmetic operations (+, -, *, /) and each number can o...
46 + 80 - 21
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [21, 80, 46], create an equation that equals 105. You can use basic arithmetic operations (+, -, *,...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5001, \"numbers\": [21, 80, 46], \"target\": 105, \"expression\": \"46 + 80 - 21\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5001, "numbers": [21, 80, 46], "target": 105, "expression": "46 + 80 - 21"}
null
null
null
null
null
null
[ "<answer>\nStep 1: Start with the number 80.\nStep 2: Add 21 to 80.\nStep 3: Subtract 46 from the result obtained in Step 2.\n</answer>\n\nLet’s verify if these steps lead us correctly to the target number:\n\n- Step 1: We start with 80.\n- Step 2: Adding 21 gives us \\(80 + 21 = 101\\).\n- Step 3: Subtracting 46 f...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, true, false, false ]
[ "80 + 21 - 46 = 105", "80 - 21 + 46", "80 + 21 + 4 = 105", "80 + 21 + 46 - 11 = 105" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 715, "span_end": 735, "original_span": "80 + 21 - 46 = 105"}, {"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 6864, "span_end": 6878, "original_span": "80 - 21 + 46"}, {"extract...
[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "80 + 21 - 46 = 105", "final_answer": 105.0}, {"is_correct": true, "error": null, "answer_block": "80 - 21 + 46", "final_answer": 105.0}, {"is_correct": false, "error": "Expression uses an unavailable or repeated number", "answer...
[ [ false, false ], [ true, false, true, true ], [ false, false ], [ false ] ]
[ [ "Step 1: Start with the number 80.\nStep 2: Add 21 to 80.\nStep 3: Subtract 46 from the result obtained in Step 2.", "80 + 21 - 46 = 105" ], [ "80 - 21 + 46", "(80 - 21) * (46 / 46) + 21 = 105", "80 - 21 + 46", "80 - 21 + 46" ], [ "80 + 21 + 4 = 105", "80 + 21 + 4 = 105" ...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 8, "span_text": "Step 1: Start with the number 80.\nStep 2: Add 21 to 80.\nStep 3: Subtract 46 from the result obtained in Step 2.", "extraction_type": "internal", "span_start": 8, "span_end": 121, "original_span": "Step 1: Start with the number 80.\nStep ...
[[{"is_correct": false, "error": "Error evaluating final answer", "answer_block": "Step 1: Start with the number 80.\nStep 2: Add 21 to 80.\nStep 3: Subtract 46 from the result obtained in Step 2.", "final_answer": null}, {"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "80 + 21...
{ "flips_by": [ 0, 1, 0, 0 ], "flips_total": 1, "num_correct": 1, "pass_at_n": 1, "percent_correct": 0.25, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, ...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776840
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [6, 76, 96], create an equation that equals 500. You can use basic arithmetic operations (+, -, *, /) and each number can on...
6*96 - 76
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [6, 76, 96], create an equation that equals 500. You can use basic arithmetic operations (+, -, *, ...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5002, \"numbers\": [6, 76, 96], \"target\": 500, \"expression\": \"6*96 - 76\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5002, "numbers": [6, 76, 96], "target": 500, "expression": "6*96 - 76"}
null
null
null
null
null
null
[ "To solve this problem using the numbers from the list [6, 76, 96] to get exactly 500 through arithmetic operations while ensuring no number repeats:\n\n### Step-by-Step Reasoning\n\n#### Initial Numbers:\n- List: [6, 76, 96]\n\n#### Goal:\n- Reach the value 500 using these numbers with allowed operations.\n\n#### ...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "Step 1: Calculate \\( 76 \\times 6 = 456 \\).\nStep 2: Add 96 to 456 resulting in \\( 456 + 96 = 552 \\).\nStep 3: Subtract 6 from 552 leading to \\( 552 - 6 = 500 \\).\n\nTherefore,\n\\( 76 \\times 6 + 96 - 6 = 500 \\)", "96 * 6 + 76 - 26 = 500", "(96 - 6) * 76 / 13.333...", "76 * 6 + 96 - 28" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 1279, "span_end": 1486, "original_span": "Step 1: Calculate \\( 76 \\times 6 = 456 \\).\nStep 2: Add 96 to 456 resulting in \\( 456 + 96 = 552 \\).\nStep 3: Subtract 6 from 552 leading to \\( 552 - 6 = 500 \\).\n\nTherefore,...
[{"is_correct": false, "error": "Could not evaluate line: Step 1: Calculate \\( 76 \\times 6 = 456 \\).", "answer_block": "Step 1: Calculate \\( 76 \\times 6 = 456 \\).\nStep 2: Add 96 to 456 resulting in \\( 456 + 96 = 552 \\).\nStep 3: Subtract 6 from 552 leading to \\( 552 - 6 = 500 \\).\n\nTherefore,\n\\( 76 \\time...
[ [ false ], [ false ], [ false ], [ false ] ]
[ [ "Step 1: Calculate \\( 76 \\times 6 = 456 \\).\nStep 2: Add 96 to 456 resulting in \\( 456 + 96 = 552 \\).\nStep 3: Subtract 6 from 552 leading to \\( 552 - 6 = 500 \\).\n\nTherefore,\n\\( 76 \\times 6 + 96 - 6 = 500 \\)" ], [ "96 * 6 + 76 - 26 = 500" ], [ "(96 - 6) * 76 / 13.333..." ], [ ...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1279, "span_text": "Step 1: Calculate \\( 76 \\times 6 = 456 \\).\nStep 2: Add 96 to 456 resulting in \\( 456 + 96 = 552 \\).\nStep 3: Subtract 6 from 552 leading to \\( 552 - 6 = 500 \\).\n\nTherefore,\n\\( 76 \\times 6 + 96 - 6 = 500 \\)", "extraction_ty...
[[{"is_correct": false, "error": "Could not evaluate line: Step 1: Calculate \\( 76 \\times 6 = 456 \\).", "answer_block": "Step 1: Calculate \\( 76 \\times 6 = 456 \\).\nStep 2: Add 96 to 456 resulting in \\( 456 + 96 = 552 \\).\nStep 3: Subtract 6 from 552 leading to \\( 552 - 6 = 500 \\).\n\nTherefore,\n\\( 76 \\tim...
{ "flips_by": [ 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 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, 0, ...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776844
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [42, 12, 3], create an equation that equals 630. You can use basic arithmetic operations (+, -, *, /) and each number can on...
42*(12 + 3)
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [42, 12, 3], create an equation that equals 630. You can use basic arithmetic operations (+, -, *, ...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5003, \"numbers\": [42, 12, 3], \"target\": 630, \"expression\": \"42*(12 + 3)\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5003, "numbers": [42, 12, 3], "target": 630, "expression": "42*(12 + 3)"}
null
null
null
null
null
null
[ "<think>\n<sample>\nFirst, I'll try to break down the process:\n\n- The goal is to achieve 630 using the numbers 42, 12, and 3 exactly once.\n- Let’s start with multiplication since larger numbers might help us get closer to 630 faster than addition or subtraction alone.\n\nStep 1: Multiply 42 and 12 together.\n\\[...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ true, true, false, true ]
[ "42 * (12 + 3)", "42 * (12 + 3)", "Step 1: Calculate 42 * 12 = 504\nStep 2: Calculate 12 + 3 = 15\nStep 3: Divide 15 by 12 to get 1.25\nStep 4: Multiply 504 by 1.25 to get 630", "Step 1: 12 + 3 = 15\nStep 2: 42 * 15 = 630" ]
[{"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 6223, "span_end": 6238, "original_span": "42 * (12 + 3)"}, {"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 5445, "span_end": 5460, "original_span": "42 * (12 + 3)"}, {"extractio...
[{"is_correct": true, "error": null, "answer_block": "42 * (12 + 3)", "final_answer": 630.0}, {"is_correct": true, "error": null, "answer_block": "42 * (12 + 3)", "final_answer": 630.0}, {"is_correct": false, "error": "Could not evaluate line: Step 1: Calculate 42 * 12 = 504", "answer_block": "Step 1: Calculate 42 * 12...
[ [ false, false, true, true ], [ true, true, true, true ], [ false, true, true, false, false ], [ true ] ]
[ [ "Step 1: 42 * 12 = 504\nStep 2: 3 * 42 = 126\nStep 3: 504 + 126 = 630", "Step 1: 42 * 12 = 504\nStep 2: 504 + 126 = 630", "42 * (12 + 3)", "42 * (12 + 3)" ], [ "42 * (12 + 3)", "Step 1: 12 + 3 = 15\nStep 2: 42 * 15 = 630", "42 * (12 + 3)", "42 * (12 + 3)" ], [ "Step 1: ...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1150, "span_text": "Step 1: 42 * 12 = 504\nStep 2: 3 * 42 = 126\nStep 3: 504 + 126 = 630", "extraction_type": "internal", "span_start": 1150, "span_end": 1218, "original_span": "Step 1: 42 * 12 = 504\nStep 2: 3 * 42 = 126\nStep 3: 504 + 126 = 630"}, {"patt...
[[{"is_correct": false, "error": "Line uses unavailable numbers: Step 2: 3 * 42 = 126", "answer_block": "Step 1: 42 * 12 = 504\nStep 2: 3 * 42 = 126\nStep 3: 504 + 126 = 630", "final_answer": 126.0}, {"is_correct": false, "error": "Line uses unavailable numbers: Step 2: 504 + 126 = 630", "answer_block": "Step 1: 42 * 1...
{ "flips_by": [ 0, 0, 1, 0 ], "flips_total": 1, "num_correct": 3, "pass_at_n": 1, "percent_correct": 0.75, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 1, 1, 2, 0 ], "reflect_close": [ 3, ...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776845
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [86, 69, 52], create an equation that equals 103. You can use basic arithmetic operations (+, -, *, /) and each number can o...
86 + 69 - 52
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [86, 69, 52], create an equation that equals 103. You can use basic arithmetic operations (+, -, *,...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5005, \"numbers\": [86, 69, 52], \"target\": 103, \"expression\": \"86 + 69 - 52\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5005, "numbers": [86, 69, 52], "target": 103, "expression": "86 + 69 - 52"}
null
null
null
null
null
null
[ "To solve this problem, we need to find a combination of the given numbers (86, 69, 52) using addition, subtraction, multiplication, or division such that their result equals 103 while ensuring each number is used exactly once.\n\nFirstly, let’s consider if there might be any straightforward way to get close to 103...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ true, true, false, false ]
[ "Step 1: Calculate the difference between 69 and 52\n69 - 52 = 17\n\nStep 2: Add the result from Step 1 to 86\n86 + 17 = 103", "Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103\n\nor simply\n\n86 + 69 - 52", "Step 1: Start with 86.\nStep 2: Add 69 to 86: 86 + 69 = 155.\nStep 3: Subtract 52 from 155: 155 - 52 = 103...
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 1486, "span_end": 1607, "original_span": "Step 1: Calculate the difference between 69 and 52\n69 - 52 = 17\n\nStep 2: Add the result from Step 1 to 86\n86 + 17 = 103"}, {"extraction_type": "final", "total_spans_found": 2, "n...
[{"is_correct": true, "error": null, "answer_block": "Step 1: Calculate the difference between 69 and 52\n69 - 52 = 17\n\nStep 2: Add the result from Step 1 to 86\n86 + 17 = 103", "final_answer": 103.0}, {"is_correct": true, "error": null, "answer_block": "Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103\n\nor simply\n\n8...
[ [ true ], [ true, true ], [ false ], [ false ] ]
[ [ "Step 1: Calculate the difference between 69 and 52\n69 - 52 = 17\n\nStep 2: Add the result from Step 1 to 86\n86 + 17 = 103" ], [ "Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103", "Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103\n\nor simply\n\n86 + 69 - 52" ], [ "Step 1: Start with 86.\nSte...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1486, "span_text": "Step 1: Calculate the difference between 69 and 52\n69 - 52 = 17\n\nStep 2: Add the result from Step 1 to 86\n86 + 17 = 103", "extraction_type": "internal", "span_start": 1486, "span_end": 1607, "original_span": "Step 1: Calculate the d...
[[{"is_correct": true, "error": null, "answer_block": "Step 1: Calculate the difference between 69 and 52\n69 - 52 = 17\n\nStep 2: Add the result from Step 1 to 86\n86 + 17 = 103", "final_answer": 103.0}], [{"is_correct": true, "error": null, "answer_block": "Step 1: 86 + 69 = 155\nStep 2: 155 - 52 = 103", "final_answe...
{ "flips_by": [ 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 2, "pass_at_n": 1, "percent_correct": 0.5, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, 1...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776847
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [63, 7, 87], create an equation that equals 354. You can use basic arithmetic operations (+, -, *, /) and each number can on...
63*7 - 87
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [63, 7, 87], create an equation that equals 354. You can use basic arithmetic operations (+, -, *, ...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5006, \"numbers\": [63, 7, 87], \"target\": 354, \"expression\": \"63*7 - 87\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5006, "numbers": [63, 7, 87], "target": 354, "expression": "63*7 - 87"}
null
null
null
null
null
null
[ "<answer>\nStep 1: Start with 63.\nStep 2: Multiply 7 by 87 which gives us 609.\nStep 3: Add the result from Step 2 (609) to the initial value (63), resulting in 672.\nStep 4: Subtract 318 from the result obtained in Step 3 (672 - 318 = 354).\n</answer>\n\nUpon reviewing, I realize my previous response did not dire...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "Step 1: Take 87 and multiply it by 4, but since we don't have 4 explicitly, we need to find another way involving our given numbers. Instead, consider breaking down the process more effectively.\nStep 2: Recognize that \\(87 \\times 4 = 348\\). This isn't directly achievable without splitting 87 into parts or find...
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 429, "span_end": 1683, "original_span": "Step 1: Take 87 and multiply it by 4, but since we don't have 4 explicitly, we need to find another way involving our given numbers. Instead, consider breaking down the process more e...
[{"is_correct": false, "error": "Could not evaluate line: Step 2: Recognize that \\(87 \\times 4 = 348\\). This isn't directly achievable without splitting 87 into parts or finding a creative combination.", "answer_block": "Step 1: Take 87 and multiply it by 4, but since we don't have 4 explicitly, we need to find anot...
[ [ false, false ], [ false ], [ false ], [ false ] ]
[ [ "Step 1: Start with 63.\nStep 2: Multiply 7 by 87 which gives us 609.\nStep 3: Add the result from Step 2 (609) to the initial value (63), resulting in 672.\nStep 4: Subtract 318 from the result obtained in Step 3 (672 - 318 = 354).", "Step 1: Take 87 and multiply it by 4, but since we don't have 4 explic...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 8, "span_text": "Step 1: Start with 63.\nStep 2: Multiply 7 by 87 which gives us 609.\nStep 3: Add the result from Step 2 (609) to the initial value (63), resulting in 672.\nStep 4: Subtract 318 from the result obtained in Step 3 (672 - 318 = 354).", "extr...
[[{"is_correct": false, "error": "Could not evaluate line: Step 4: Subtract 318 from the result obtained in Step 3 (672 - 318 = 354).", "answer_block": "Step 1: Start with 63.\nStep 2: Multiply 7 by 87 which gives us 609.\nStep 3: Add the result from Step 2 (609) to the initial value (63), resulting in 672.\nStep 4: Su...
{ "flips_by": [ 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 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, 0, ...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776848
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>. # Problem Using the numbers in the list [8, 50, 19], create an equation that equals 102. You can use basic arithmetic operations (+, -, *, /) and each number can on...
19*8 - 50
countdown_3arg
countdown
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(your answer)</answer>.\n\n# Problem\nUsing the numbers in the list [8, 50, 19], create an equation that equals 102. You can use basic arithmetic operations (+, -, *, ...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 5007, \"numbers\": [8, 50, 19], \"target\": 102, \"expression\": \"19*8 - 50\"}", "task_source": "countdown"}
test
null
null
null
null
null
null
null
null
null
null
null
null
{"source_dataset": "countdown", "source_index": 5007, "numbers": [8, 50, 19], "target": 102, "expression": "19*8 - 50"}
null
null
null
null
null
null
[ "To solve the given problem using the numbers from the list [8, 50, 19] to make 102, we need to consider various combinations involving addition, subtraction, multiplication, and division while ensuring each number is used exactly once.\n\nFirstly, let’s explore if there’s any straightforward combination:\n\n- Mult...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.36.208:9080/v1", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 16384, "n": 1, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vl...
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "50 + 19 + 33", "50 + 8 + (19 * 2 - 19) = 102", "50 + 19 + (8 * 4)", "Step 1: 19 / 19 + 1 = 2\nStep 2: 50 * 2 = 100\nStep 3: 100 + 2 = 102" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 1868, "span_end": 1882, "original_span": "50 + 19 + 33"}, {"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 634, "span_end": 664, "original_span": "50 + 8 + (19 * 2 - 19) = 102"},...
[{"is_correct": false, "error": "Expression uses an unavailable or repeated number", "answer_block": "50 + 19 + 33", "final_answer": 102.0}, {"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "50 + 8 + (19 * 2 - 19) = 102", "final_answer": 102.0}, {"is_correct": false, "error": "E...
[ [ false ], [ false ], [ false, false ], [ false, false, false, false ] ]
[ [ "50 + 19 + 33" ], [ "50 + 8 + (19 * 2 - 19) = 102" ], [ "50 + 19 + 33", "50 + 19 + (8 * 4)" ], [ "50 + 19 + 8 * 2 - 6", "50 + 19 + 33 = 102", "Step 1: 19 / 19 + 1 = 2\nStep 2: 50 * 2 = 100\nStep 3: 100 + 2 = 102", "Step 1: 19 / 19 + 1 = 2\nStep 2: 50 * 2 = 100\nStep 3: ...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1868, "span_text": "50 + 19 + 33", "extraction_type": "internal", "span_start": 1868, "span_end": 1882, "original_span": "50 + 19 + 33"}], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 634, "span_text": "50 + 8 + (19 * 2 - 19) = 102", "ext...
[[{"is_correct": false, "error": "Expression uses an unavailable or repeated number", "answer_block": "50 + 19 + 33", "final_answer": 102.0}], [{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "50 + 8 + (19 * 2 - 19) = 102", "final_answer": 102.0}], [{"is_correct": false, "error...
{ "flips_by": [ 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 ], "best_of_n": [ 0, 0, 0, 2 ], "reflect_close": [ 0, 0, ...
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776850
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
40 + 46 + 38
countdown_3arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5008, \\\"number(...TRUNCATED)
test
null
null
null
null
null
null
null
null
null
null
null
null
"{\"source_dataset\": \"countdown\", \"source_index\": 5008, \"numbers\": [38, 46, 40], \"target\": (...TRUNCATED)
null
null
null
null
null
null
["<answer>\nStep 1: Start with 46.\nStep 2: Add 38 to 46.\nStep 3: Subtract 40 from the result of St(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.36.208:9080/v1","backend":"litellm","chat_template_applied":true,"generatio(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, true, true, true ]
["(46 + 38 - 40)","38 + 46 + 40 = 124","Step 1: 46 + 40 = 86\nStep 2: 86 + 38 = 124","Step 1: 46 + 3(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 2, \"non_overlapping_spans\": 2, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\": (...TRUNCATED)
[ [ false, false ], [ true ], [ true ], [ false, true, true ] ]
[["Step 1: Start with 46.\nStep 2: Add 38 to 46.\nStep 3: Subtract 40 from the result of Step 2.","((...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 8, \"span_text\": \"Step 1:(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Error evaluating final answer\", \"answer_block\": \"Step 1:(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":3,"pass_at_n":1,"percent_correct":0.75,"skill_co(...TRUNCATED)
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776855
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
19 + 92 + 21
countdown_3arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5009, \\\"number(...TRUNCATED)
test
null
null
null
null
null
null
null
null
null
null
null
null
"{\"source_dataset\": \"countdown\", \"source_index\": 5009, \"numbers\": [92, 21, 19], \"target\": (...TRUNCATED)
null
null
null
null
null
null
["<answer>\nStep 1: Start with 92.\nStep 2: Add 21 to 92: 92 + 21 = 113.\nStep 3: Add 19 to 113: 113(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.36.208:9080/v1","backend":"litellm","chat_template_applied":true,"generatio(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, true, true, true ]
["Step 1: Start with 92.\nStep 2: Add 21 to 92: 92 + 21 = 113.\nStep 3: Add 19 to 113: 113 + 19 = 13(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 1, \"non_overlapping_spans\": 1, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Could not evaluate line: Step 2: Add 21 to 92: 92 + 21 = 113.(...TRUNCATED)
[ [ false ], [ true ], [ true, true, true, true ], [ true, true, true, true, true, true ] ]
[["Step 1: Start with 92.\nStep 2: Add 21 to 92: 92 + 21 = 113.\nStep 3: Add 19 to 113: 113 + 19 = 1(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 8, \"span_text\": \"Step 1:(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Could not evaluate line: Step 2: Add 21 to 92: 92 + 21 = 113(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":3,"pass_at_n":1,"percent_correct":0.75,"skill_co(...TRUNCATED)
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776857
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
23*(21 - 7)
countdown_3arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5010, \\\"number(...TRUNCATED)
test
null
null
null
null
null
null
null
null
null
null
null
null
"{\"source_dataset\": \"countdown\", \"source_index\": 5010, \"numbers\": [23, 21, 7], \"target\": 3(...TRUNCATED)
null
null
null
null
null
null
["<answer>\nStep 1: Start with the largest number, 23.\nStep 2: Multiply 23 by 7 to get 161.\nStep 3(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.36.208:9080/v1","backend":"litellm","chat_template_applied":true,"generatio(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/SkillFactory/M-OT_ours_1k_SFT-sft", "n_responses": 4 }
[ false, false, false, false ]
["Step 1: Start with the largest number, 23.\nStep 2: Multiply 23 by 7 to get 161.\nStep 3: Add 21 t(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 1, \"non_overlapping_spans\": 1, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Could not evaluate line: Step 6: Calculate (23 * 7) + 21 = 16(...TRUNCATED)
[ [ false ], [ false ], [ false ], [ false ] ]
[["Step 1: Start with the largest number, 23.\nStep 2: Multiply 23 by 7 to get 161.\nStep 3: Add 21 (...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 8, \"span_text\": \"Step 1:(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Could not evaluate line: Step 6: Calculate (23 * 7) + 21 = 1(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-18
test
main
SkillFactory/M-OT_ours_1k_SFT-sft
0
eval_sft
1
2025-11-18T22:20:05.776859
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_OT1k_Ours-SFT-eval_sft
End of preview. Expand in Data Studio

Experiment Tracker: FinEval_16k_fulleval_OT1k_Ours-SFT

Experiment Description: Simple test experiment for Skill Factory workflows.

Start Time: 2025-11-18T17:56:31.047037

Tracker Dataset: TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1

Stages Completed

Total stages: 1

Models Created

Dataset Configurations

This tracker dataset contains the following configurations with immediate upload as stages complete:

Training Data (Complete Datasets)

Hyperparameters (Complete Configurations)

Logs (Stage-Specific)

Evaluation Results (Complete with Annotations)

Metadata

  • experiment_metadata: Timeline and stage information

Usage

Load specific configurations with:

from datasets import load_dataset

# Load experiment metadata
metadata = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'experiment_metadata')

# Load complete training datasets
sft_data = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'training_data__sft')
sft_metadata = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'training_data__sft_metadata')

# Load complete configurations
sft_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'hyperparameters__sft')
rl_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'hyperparameters__rl')

# Load stage-specific logs
sft_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'logs__sft')
rl_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'logs__rl')

# Load evaluation results with annotations
sft_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'evals_eval_sft')
rl_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_OT1k_Ours-SFT__v1', 'evals_eval_rl')

Models

Registry

All models from this experiment are automatically registered in the SkillFactory Model Registry with:

  • Complete training configuration (hyperparameters, datasets, methods)
  • Experiment lineage (links back to this tracker dataset)
  • Stage-specific metadata (SFT vs RL training details)
  • Structured input data references (training datasets and configurations)

Registry entries follow the naming pattern: Model - FinEval_16k_fulleval_OT1k_Ours-SFT - {stage_name} - {SFT/RL}


Generated by SkillFactory Experiment Management System All artifacts uploaded immediately as stages complete with perfect data provenance

Downloads last month
4