Dataset Viewer
Auto-converted to Parquet Duplicate
question
stringlengths
984
989
answer
stringlengths
17
28
task_config
stringclasses
1 value
task_source
stringclasses
1 value
prompt
listlengths
1
1
model_responses
listlengths
0
0
model_responses__eval_is_correct
listlengths
0
0
all_other_columns
stringlengths
192
210
original_split
stringclasses
1 value
metadata
stringlengths
134
152
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
379
3.27k
model_responses__best_of_n_atags__eval_evaluation_metadata
stringlengths
397
3.17k
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
1.35k
82.5k
model_responses__best_of_n_atags__internal_answers__eval_evaluation_metadata
stringlengths
831
37.9k
model_responses__best_of_n_atags__metrics
dict
eval_date
stringdate
2025-11-10 00:00:00
2025-11-10 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
2025-11-10 08:13:59
2025-11-10 08:13:59
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 [37, 27, 3, 84, 5, 80], create an equation that equals 152. You can use basic arithmetic operations (+, -, *, /) and each nu...
84 + 80 - 37 + 3 - 5 + 27
countdown_6arg
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 [37, 27, 3, 84, 5, 80], create an equation that equals 152. You can use basic arithmetic operations...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 2, \"numbers\": [37, 27, 3, 84, 5, 80], \"target\": 152, \"expression\": \"84 + 80 - 37 + 3 - 5 + 27\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 2, "numbers": [37, 27, 3, 84, 5, 80], "target": 152, "expression": "84 + 80 - 37 + 3 - 5 + 27"}
[ "<think>\n<sample>\nRewritten Question:\n\nUse the numbers from the given set {37, 27, 3, 84, 5, 80} exactly once to form an equation equaling 152 using addition, subtraction, multiplication, and division.\n\nStep-by-step thinking process:\n\n1. We need to find a combination of these numbers that adds up to 152.\n2...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.26:9080/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_vll...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "\\( 120 + 4 = 124 \\\\\n124 + 8 = 132 \\\\\n132 + 10 = 142 \\\\\n142 + 10 = 152\n\\)", "(84 - 37) + (27 + 3) + 80 - 15", "80 - 3 = 77 * 2", "None of the above" ]
[{"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 4303, "span_end": 4377, "original_span": "\\( 120 + 4 = 124 \\\\\n124 + 8 = 132 \\\\\n132 + 10 = 142 \\\\\n142 + 10 = 152\n\\)"}, {"extraction_type": "final", "total_spans_found": 10, "non_overlapping_spans": 10, "span_start...
[{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "\\( 120 + 4 = 124 \\\\\n124 + 8 = 132 \\\\\n132 + 10 = 142 \\\\\n142 + 10 = 152\n\\)", "final_answer": null}, {"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(84 - 37) + (27 + 3) + 80 - 15", "fin...
[ [ false, false, false ], [ false, false, false, false, false, false, false, false, false, false ], [ false, false, false, false, false, false ], [ false, false, false ] ]
[ [ "\\( 84 + 68 = 152 \\)", "\\( 120 + 4 = 124 \\\\\n124 + 8 = 132 \\\\\n132 + 10 = 142 \\\\\n142 + 10 = 152\n\\)", "\\( 120 + 4 = 124 \\\\\n124 + 8 = 132 \\\\\n132 + 10 = 142 \\\\\n142 + 10 = 152\n\\)" ], [ "No specific unique sequence found meeting criteria accurately within standard arithmetic...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1082, "span_text": "\\( 84 + 68 = 152 \\)", "extraction_type": "internal", "span_start": 1082, "span_end": 1104, "original_span": "\\( 84 + 68 = 152 \\)"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 3883, "span_text": "\\( 120 + 4 = 124 ...
[[{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "\\( 84 + 68 = 152 \\)", "final_answer": null}, {"is_correct": false, "error": "Could not evaluate expression", "answer_block": "\\( 120 + 4 = 124 \\\\\n124 + 8 = 132 \\\\\n132 + 10 = 142 \\\\\n142 + 10 = 152\n\\)", "final_answer": null},...
{ "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, 2, 1, 0 ], "reflect_close": [ 2, 8, ...
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568438
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
20*(64 + 48 - 27 - 45) - 81
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 3, \\\"numbers\\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 3, \"numbers\": [27, 48, 64, 20, 81, 45], \"t(...TRUNCATED)
["<think>\n<sample>\nRewritten Question:\n\nCreate an equation using the numbers from the given set (...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["(64 + 81) + (27 + 48) + (20 + 45) + 1","(27 + 48) * 64 + 20 + 81 + 45","(48 + 64 + 20 + 45)","((27(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 11, \"non_overlapping_spans\": 11, \"span_(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\": (...TRUNCATED)
[[false,false,false,false,false,false,false,false,false,false,false],[false,false,false,false,false,(...TRUNCATED)
[["(81 * 45) - 27","\\(27 + 48 + 64 - 20 - 81 + 45 = 719\\)","\\(27 + 48 + 64 + 20 + 81 + 45 = 719\\(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 1003, \"span_text\": \"(81 (...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\":(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568447
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
9*56 - 71 - 59 + 74 + 44
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 5, \\\"numbers\\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 5, \"numbers\": [74, 44, 56, 9, 59, 71], \"ta(...TRUNCATED)
["<think>\n<sample>\nTo solve for 492 using the given numbers [74, 44, 56, 9, 59, 71], we need to fi(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ null, "(74 * 56 - 9 + 59) / 71", "(74 + 56) * 9 - 59 / 56 + 44 = 492", null ]
"[{\"extraction_type\": \"final\"}, {\"extraction_type\": \"final\", \"total_spans_found\": 11, \"no(...TRUNCATED)
"[{\"evaluation_method\": \"response_processor\", \"reason\": \"no_extracted_answer\", \"extraction_(...TRUNCATED)
[ [], [ false, false, false, false, false, false, false, false, false, false, false ], [ false, false, false ], [] ]
[[],["\\( 74 \\times 71 - 59 + 44 + \\frac{5079}{9} \\)","\\( 74 \\times 71 - 59 + 44 + \\frac{5079}(...TRUNCATED)
"[[], [{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 1870, \"span_text\": \"(...TRUNCATED)
"[[], [{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"\\\(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568449
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
27*(75 - 52 - 17 + 37 - 33)
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 16, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 16, \"numbers\": [75, 17, 52, 33, 37, 27], \"(...TRUNCATED)
["<think>\n<sample>\nRewritten Question:\n\nCreate an equation using the given numbers [75, 17, 52, (...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["No specific numerical outcome found within feasible limits.","(75 - 33) * 52 / 37 + 27","\\( 75 - (...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 2, \"non_overlapping_spans\": 2, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"No speci(...TRUNCATED)
[[false,false],[false,false,false,false,false,false],[false,false],[false,false,false,false,false,fa(...TRUNCATED)
[["((75 * 52) - (33 * 37)) + 27 = 270","No specific numerical outcome found within feasible limits."(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 551, \"span_text\": \"((75 (...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\":(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568450
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
27*(52 - 12 - 37 + 4 + 10)
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 19, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 19, \"numbers\": [27, 4, 37, 10, 12, 52], \"t(...TRUNCATED)
["<think>\n<sample>\nRewritten Question:\n\nCreate an equation using the given numbers [27, 4, 37, 1(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "27 + 12 - 8 = 459", "((27 * 12) + 4 - 37) / 10 + 52", "(52 - 10 + 37 * 6 - 12)", "27 * 17 = 459" ]
"[{\"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 ], [ false ], [ false, false, false, false, false, false, false ], [ false, false, false, false, false ] ]
[["27 + 12 - 8 = 459","27 + 12 - 8 = 459"],["((27 * 12) + 4 - 37) / 10 + 52"],["(x * 52 + 21) / 37",(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 869, \"span_text\": \"27 + (...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\":(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568451
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
34*22 - 67 - 84 + 3 - 55
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 20, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 20, \"numbers\": [3, 55, 22, 34, 84, 67], \"t(...TRUNCATED)
["<think>\n<sample>\nRewritten Question:\n\nCreate an equation using the given numbers [3, 55, 22, 3(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["84 * 67 - 3 - 55 - 22 - 34","(84 * 6) + ((55 - 34)) - 19 + 39","(84 + 67) * 3 + 55 - 34 + 22 - 19"(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 12, \"non_overlapping_spans\": 12, \"span_(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\": (...TRUNCATED)
[[false,false,false,false,false,false,false,false,false,false,false,false],[false,false],[false,fals(...TRUNCATED)
[["(84 * (55 - 22)) + 34 = 545","Step 1: 55 * 9 = 495\nStep 2: 495 + 50 = 545","\\( 34 \\times 17 - (...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 342, \"span_text\": \"(84 *(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\":(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568453
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
52*(16 + 81 - 79) - 69 - 88
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 22, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 22, \"numbers\": [81, 79, 88, 69, 52, 16], \"(...TRUNCATED)
["<think>\n<sample>\nRewritten Question:\n\nCreate an equation using the numbers from the given set (...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["No valid solution found.","(81 + 79) + 88 - 69 + 52 + 16","((88 + 69) * (16 - 52/81))","\\(88 \\ti(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 3, \"non_overlapping_spans\": 3, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"No valid(...TRUNCATED)
[[false,false,false],[false,false,false,false,false,false,false,false,false,false,false,false,false](...TRUNCATED)
[["\\( 81 + 79 + 88 + 69 - 157 \\times 48 = 779 \\)","None found.","No valid solution found."],["(77(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 1110, \"span_text\": \"\\\\(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"\\\\( 8(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568454
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
28*(37 - 19 + 26 + 19 - 40)
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 27, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 27, \"numbers\": [19, 19, 37, 26, 40, 28], \"(...TRUNCATED)
["<think>\n<sample>\nTo solve for 644 using the given numbers [19, 19, 37, 26, 40, 28] while ensurin(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["(37 + 28 - 19) + 40 + 26","\\( (19 \\times 37) + 26 - 28 = 644 \\)","Step 1: 40 * 16 = 640\nStep 2(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 12, \"non_overlapping_spans\": 12, \"span_(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\": (...TRUNCATED)
[[false,false,false,false,false,false,false,false,false,false,false,false],[false,false,false,false,(...TRUNCATED)
[["\\(37 + 40 + 19 + 26 + 28 = 150\\)","629 + 28","(19 * 37) - (19 * 28) + (19 * 26)","Step 1: 40 * (...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 2454, \"span_text\": \"\\\\(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"\\\\(37(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568455
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
-2 + (84*22 - 25 + 59)/2
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 29, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 29, \"numbers\": [25, 59, 2, 84, 22, 2], \"ta(...TRUNCATED)
["<think>\n<sample>\nTo solve for the target number using the given set of numbers [25, 59, 2, 84, 2(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["25 * 37 + 14 = 939","(59 * 18) + 25 - 22 + 22 - 22 + 22 + 22 + 22 + 22","(84 - 59) * 22 + 2","\\( (...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 8, \"non_overlapping_spans\": 8, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression uses an unavailable or repeated number\", \"answer(...TRUNCATED)
[[false,false,false,false,false,false,false,false],[false,false,false,false,false,false,false,false,(...TRUNCATED)
[["No valid sequence found under specified constraints leading to exact 939.","(25 * 37) + 22 - 59",(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 2561, \"span_text\": \"No v(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"No vali(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568456
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
4*(49 - 19 + 82) + 89 - 29
countdown_6arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 30, \\\"numbers\(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 30, \"numbers\": [4, 49, 19, 89, 29, 82], \"t(...TRUNCATED)
["<think>\n<sample>\nFirst, I'll try to understand the problem better by writing out a plan and go r(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.26:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_ours_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["No such equation exists using all five numbers exactly once.","\\( (4 + 49) * 19 + 89 - 29 = 508 \(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 3, \"non_overlapping_spans\": 3, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Could not evaluate expression\", \"answer_block\": \"No such (...TRUNCATED)
[[false,false,false],[false,false,false],[false,false,false,false,false,false],[false,false,false,fa(...TRUNCATED)
[["(89 - 4) * (49 + 19) / 29 - 82","No such equation exists using all five numbers exactly once.","N(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 908, \"span_text\": \"(89 -(...TRUNCATED)
"[[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\":(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"skill_cou(...TRUNCATED)
2025-11-10
test
main
TAUR-dev/M-AT_ours_sft-sft
0
eval_sft
1
2025-11-10T08:13:59.568458
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg-eval_sft
End of preview. Expand in Data Studio

Experiment Tracker: FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg

Experiment Description: Evaluation experiment for task countdown_6arg from FinEval_16k_fulleval_AT_OURS-SFT

Start Time: 2025-11-10T06:12:59.975896

Tracker Dataset: TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__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_AT_OURS-SFT-countdown_6arg__v1', 'experiment_metadata')

# Load complete training datasets
sft_data = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'training_data__sft')
sft_metadata = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'training_data__sft_metadata')

# Load complete configurations
sft_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'hyperparameters__sft')
rl_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'hyperparameters__rl')

# Load stage-specific logs
sft_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'logs__sft')
rl_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'logs__rl')

# Load evaluation results with annotations
sft_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__v1', 'evals_eval_sft')
rl_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_OURS-SFT-countdown_6arg__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_AT_OURS-SFT-countdown_6arg - {stage_name} - {SFT/RL}


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

Downloads last month
3