Dataset Viewer
Auto-converted to Parquet Duplicate
question
stringlengths
980
985
answer
stringlengths
13
23
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
186
201
original_split
stringclasses
1 value
metadata
stringlengths
128
143
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
238
2.32k
model_responses__best_of_n_atags__eval_evaluation_metadata
stringlengths
402
2.27k
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
205
4.23k
model_responses__best_of_n_atags__internal_answers__eval_evaluation_metadata
stringlengths
119
2.28k
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 01:46:55
2025-11-10 01:46:55
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 [68, 97, 78, 16, 66], create an equation that equals 137. You can use basic arithmetic operations (+, -, *, /) and each numb...
97 - 16 - 78 + 66 + 68
countdown_5arg
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 [68, 97, 78, 16, 66], create an equation that equals 137. You can use basic arithmetic operations (...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 250, \"numbers\": [68, 97, 78, 16, 66], \"target\": 137, \"expression\": \"97 - 16 - 78 + 66 + 68\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 250, "numbers": [68, 97, 78, 16, 66], "target": 137, "expression": "97 - 16 - 78 + 66 + 68"}
[ "To solve for 137 using the given numbers [68, 97, 78, 16, 66], we need to find a combination of these numbers through addition, subtraction, multiplication, and division that will result in 137. Let's break down the process:\n\n### Step-by-Step Solution\n\n**Step 1:** Start with one of the larger numbers.\n - We...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "\\( 97 - 68 + 78 \\)", "(97 - 66) + 68 + 78 - 40", "((68 + 97) - 78) + 16", "(68 + 66) * 2 - 97 + 78 = 137" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 2455, "span_end": 2476, "original_span": "\\( 97 - 68 + 78 \\)"}, {"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 641, "span_end": 668, "original_span": "(97 - 66) + 68 + 78 - 4...
[{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "\\( 97 - 68 + 78 \\)", "final_answer": null}, {"is_correct": false, "error": "Expression uses an unavailable or repeated number", "answer_block": "(97 - 66) + 68 + 78 - 40", "final_answer": 137.0}, {"is_correct": false, "error": "Expressi...
[ [ false ], [ false ], [ false ], [ false ] ]
[ [ "\\( 97 - 68 + 78 \\)" ], [ "(97 - 66) + 68 + 78 - 40" ], [ "((68 + 97) - 78) + 16" ], [ "(68 + 66) * 2 - 97 + 78 = 137" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 2455, "span_text": "\\( 97 - 68 + 78 \\)", "extraction_type": "internal", "span_start": 2455, "span_end": 2476, "original_span": "\\( 97 - 68 + 78 \\)"}], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 641, "span_text": "(97 - 66) + 68 + 78...
[[{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "\\( 97 - 68 + 78 \\)", "final_answer": null}], [{"is_correct": false, "error": "Expression uses an unavailable or repeated number", "answer_block": "(97 - 66) + 68 + 78 - 40", "final_answer": 137.0}], [{"is_correct": false, "error": "Exp...
{ "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, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860917
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-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 [34, 81, 40, 25, 24], create an equation that equals 106. You can use basic arithmetic operations (+, -, *, /) and each numb...
34 + 81 - 24 - 25 + 40
countdown_5arg
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 [34, 81, 40, 25, 24], create an equation that equals 106. You can use basic arithmetic operations (...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 251, \"numbers\": [34, 81, 40, 25, 24], \"target\": 106, \"expression\": \"34 + 81 - 24 - 25 + 40\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 251, "numbers": [34, 81, 40, 25, 24], "target": 106, "expression": "34 + 81 - 24 - 25 + 40"}
[ "Rewritten Question:\n\nCreate an equation using the numbers from the given set {34, 81, 40, 25, 24} exactly once such that the result equals 106.\n\nStep 1: First, we need to find two numbers whose sum will help us get close to 106.\n - Let’s try adding 81 and 25 because their total is already quite large (81 + ...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ true, false, true, true ]
[ "81 + 25 = 106", null, "(81 + 25)", "(81 + 25)" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 346, "span_end": 359, "original_span": "81 + 25 = 106"}, {"extraction_type": "final"}, {"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 715, "span_end": 727, "original_span": "(8...
[{"is_correct": true, "error": null, "answer_block": "81 + 25 = 106", "final_answer": 106.0}, {"evaluation_method": "response_processor", "reason": "no_extracted_answer", "extraction_result": "{'extraction_type': 'final'}"}, {"is_correct": true, "error": null, "answer_block": "(81 + 25)", "final_answer": 106.0}, {"is_c...
[ [ true ], [], [ true ], [ true ] ]
[ [ "81 + 25 = 106" ], [], [ "(81 + 25)" ], [ "(81 + 25)" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 346, "span_text": "81 + 25 = 106", "extraction_type": "internal", "span_start": 346, "span_end": 359, "original_span": "81 + 25 = 106"}], [], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 715, "span_text": "(81 + 25)", "extraction_type": "...
[[{"is_correct": true, "error": null, "answer_block": "81 + 25 = 106", "final_answer": 106.0}], [], [{"is_correct": true, "error": null, "answer_block": "(81 + 25)", "final_answer": 106.0}], [{"is_correct": true, "error": null, "answer_block": "(81 + 25)", "final_answer": 106.0}]]
{ "flips_by": [ 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 3, "pass_at_n": 1, "percent_correct": 0.75, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860930
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-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 [100, 93, 82, 18, 63], create an equation that equals 903. You can use basic arithmetic operations (+, -, *, /) and each num...
18*(63 - 100 + 82) + 93
countdown_5arg
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 [100, 93, 82, 18, 63], create an equation that equals 903. You can use basic arithmetic operations ...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 252, \"numbers\": [100, 93, 82, 18, 63], \"target\": 903, \"expression\": \"18*(63 - 100 + 82) + 93\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 252, "numbers": [100, 93, 82, 18, 63], "target": 903, "expression": "18*(63 - 100 + 82) + 93"}
[ "Rewritten Question:\n\nCreate an equation using the numbers from the given set [100, 93, 82, 18, 63] exactly once so that the result equals 903.\n\nStep 1: Start with one of the largest numbers.\n - We'll begin with 100 because it's close to our target (903).\n\nStep 2: Subtract another large number from 100.\n ...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "(100 - 93) + 82 + 63", "100 + 93 - (82 - 18) + 63", null, null ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 951, "span_end": 973, "original_span": "(100 - 93) + 82 + 63"}, {"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 406, "span_end": 431, "original_span": "100 + 93 - (82 - 18) + 63...
[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(100 - 93) + 82 + 63", "final_answer": 903.0}, {"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "100 + 93 - (82 - 18) + 63", "final_answer": 903.0}, {"evaluation_method": "response_proces...
[ [ false ], [ false ], [], [] ]
[ [ "(100 - 93) + 82 + 63" ], [ "100 + 93 - (82 - 18) + 63" ], [], [] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 951, "span_text": "(100 - 93) + 82 + 63", "extraction_type": "internal", "span_start": 951, "span_end": 973, "original_span": "(100 - 93) + 82 + 63"}], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 406, "span_text": "100 + 93 - (82 - 18) +...
[[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(100 - 93) + 82 + 63", "final_answer": 903.0}], [{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "100 + 93 - (82 - 18) + 63", "final_answer": 903.0}], [], []]
{ "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, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860933
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-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 [49, 4, 7, 68, 77], create an equation that equals 716. You can use basic arithmetic operations (+, -, *, /) and each number...
4*7*(77 - 49) - 68
countdown_5arg
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 [49, 4, 7, 68, 77], create an equation that equals 716. You can use basic arithmetic operations (+,...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 253, \"numbers\": [49, 4, 7, 68, 77], \"target\": 716, \"expression\": \"4*7*(77 - 49) - 68\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 253, "numbers": [49, 4, 7, 68, 77], "target": 716, "expression": "4*7*(77 - 49) - 68"}
[ "Rewritten Question:\n\nCreate an equation using the numbers from the given list [49, 4, 7, 68, 77] that equals 716. Each number must be used exactly once.\n \nTo solve this, we need to find a combination of these numbers through addition, subtraction, multiplication, division, and parentheses so that their sum eq...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "No valid equation found.", "(4 * 77) - (68 - 4) = 716", "(77 * 9 = 716)", "(49 * 13 - 7) + 68 + 77 = 716" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 2523, "span_end": 2547, "original_span": "No valid equation found."}, {"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 364, "span_end": 391, "original_span": "(4 * 77) - (68 - 4)...
[{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "No valid equation found.", "final_answer": null}, {"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(4 * 77) - (68 - 4) = 716", "final_answer": 716.0}, {"is_correct": false, "error": "Could not eva...
[ [ false ], [ false ], [ false ], [ false ] ]
[ [ "No valid equation found." ], [ "(4 * 77) - (68 - 4) = 716" ], [ "(77 * 9 = 716)" ], [ "(49 * 13 - 7) + 68 + 77 = 716" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 2523, "span_text": "No valid equation found.", "extraction_type": "internal", "span_start": 2523, "span_end": 2547, "original_span": "No valid equation found."}], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 364, "span_text": "(4 * 77) - ...
[[{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "No valid equation found.", "final_answer": null}], [{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(4 * 77) - (68 - 4) = 716", "final_answer": 716.0}], [{"is_correct": false, "error": "Could no...
{ "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, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860934
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-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 [96, 6, 42, 93, 52], create an equation that equals 103. You can use basic arithmetic operations (+, -, *, /) and each numbe...
96 + 52 + 6 - 93 + 42
countdown_5arg
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, 6, 42, 93, 52], create an equation that equals 103. You can use basic arithmetic operations (+...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 255, \"numbers\": [96, 6, 42, 93, 52], \"target\": 103, \"expression\": \"96 + 52 + 6 - 93 + 42\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 255, "numbers": [96, 6, 42, 93, 52], "target": 103, "expression": "96 + 52 + 6 - 93 + 42"}
[ "Rewritten Question:\n\nCreate an equation using the given numbers [96, 6, 42, 93, 52] which equals 103.\n- Step 1: Add 96 and 42 together first because their sum will likely lead us closer to our goal of 103 quickly.\n \n<answer>\nStep 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "Step 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since subtracting 93 from 138 gives us something close but not exactly 103 yet.\nStep 3: Finally add 52 to get the exact value of 103.", "\\( \\frac{96}{6} + \\frac{42*93}{52} \\)", "((96 - 6) + 93 + 52)", "((96 - 6) * (42 + 93)) -...
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 225, "span_end": 437, "original_span": "Step 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since subtracting 93 from 138 gives us something close but not exactly 103 yet.\nStep 3: Finally add 52 to...
[{"is_correct": false, "error": "Incorrect final answer", "answer_block": "Step 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since subtracting 93 from 138 gives us something close but not exactly 103 yet.\nStep 3: Finally add 52 to get the exact value of 103.", "final_answer": 138.0}, {"is_c...
[ [ false ], [ false ], [ false ], [ false ] ]
[ [ "Step 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since subtracting 93 from 138 gives us something close but not exactly 103 yet.\nStep 3: Finally add 52 to get the exact value of 103." ], [ "\\( \\frac{96}{6} + \\frac{42*93}{52} \\)" ], [ "((96 - 6) + 93 + 52)" ...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 225, "span_text": "Step 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since subtracting 93 from 138 gives us something close but not exactly 103 yet.\nStep 3: Finally add 52 to get the exact value of 103.", "extraction_type": "in...
[[{"is_correct": false, "error": "Incorrect final answer", "answer_block": "Step 1: 96 + 42 = 138\nStep 2: Subtract 93 from the result obtained in Step 1 since subtracting 93 from 138 gives us something close but not exactly 103 yet.\nStep 3: Finally add 52 to get the exact value of 103.", "final_answer": 138.0}], [{"i...
{ "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, 1 ], "reflect_close": [ 0, 0, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860936
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-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 [91, 36, 19, 21, 66], create an equation that equals 932. You can use basic arithmetic operations (+, -, *, /) and each numb...
21*36 + 19 + 91 + 66
countdown_5arg
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 [91, 36, 19, 21, 66], create an equation that equals 932. You can use basic arithmetic operations (...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 257, \"numbers\": [91, 36, 19, 21, 66], \"target\": 932, \"expression\": \"21*36 + 19 + 91 + 66\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 257, "numbers": [91, 36, 19, 21, 66], "target": 932, "expression": "21*36 + 19 + 91 + 66"}
[ "First, I'll try to understand the problem better by writing out a plan and go really deep into detail about how I should solve this.\nThe goal is to create an equation using the given numbers [91, 36, 19, 21, 66] that equals 932. Each number can only be used once, so we need to find a combination of these numbers ...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "(91 - 66) + (19 + 21) + (36)", null, "No valid solution found under strict integer arithmetic constraints.", "\\(91 + 36 - 19 * 9 + 10\\)" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 1577, "span_end": 1608, "original_span": "(91 - 66) + (19 + 21) + (36)"}, {"extraction_type": "final"}, {"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 3117, "span_end": 3187, "...
[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(91 - 66) + (19 + 21) + (36)", "final_answer": 932.0}, {"evaluation_method": "response_processor", "reason": "no_extracted_answer", "extraction_result": "{'extraction_type': 'final'}"}, {"is_correct": false, "error": "Could not ...
[ [ false ], [], [ false ], [ false ] ]
[ [ "(91 - 66) + (19 + 21) + (36)" ], [], [ "No valid solution found under strict integer arithmetic constraints." ], [ "\\(91 + 36 - 19 * 9 + 10\\)" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1577, "span_text": "(91 - 66) + (19 + 21) + (36)", "extraction_type": "internal", "span_start": 1577, "span_end": 1608, "original_span": "(91 - 66) + (19 + 21) + (36)"}], [], [{"pattern_used": "answer tag", "confidence": 1.0, "position": 3117, "span_text":...
[[{"is_correct": false, "error": "Expression does not equal stated result", "answer_block": "(91 - 66) + (19 + 21) + (36)", "final_answer": 932.0}], [], [{"is_correct": false, "error": "Could not evaluate expression", "answer_block": "No valid solution found under strict integer arithmetic constraints.", "final_answer"...
{ "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, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860938
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-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 [57, 72, 1, 25, 12], create an equation that equals 165. You can use basic arithmetic operations (+, -, *, /) and each numbe...
57 - 1 + 72 + 12 + 25
countdown_5arg
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 [57, 72, 1, 25, 12], create an equation that equals 165. You can use basic arithmetic operations (+...
[]
[]
{"metadata": "{\"source_dataset\": \"countdown\", \"source_index\": 258, \"numbers\": [57, 72, 1, 25, 12], \"target\": 165, \"expression\": \"57 - 1 + 72 + 12 + 25\"}", "task_source": "countdown"}
test
{"source_dataset": "countdown", "source_index": 258, "numbers": [57, 72, 1, 25, 12], "target": 165, "expression": "57 - 1 + 72 + 12 + 25"}
[ "Rewritten Question:\n\nCreate an equation using the numbers from the given set {57, 72, 1, 25, 12} that equals 165. Each number must be used exactly once.\n\nStep-by-step thinking process:\n\n1. First, let's consider adding some combination of these numbers together.\n - We need to get close to 165 but not excee...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.10:10080/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/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, true, false ]
[ "Step 1: 72 + 12 = 84 \nStep 2: 84 + 57 = 141 \nStep 3: 141 - 114 = 27 \nStep 4: 27 * 6 = 162 \nStep 5: 162 + 3 = 165", "\\(165 = 12 + 25 + 1 + 72 - 57\\)", "((57 + 72) - 1) + 25 + 12 = 165", "\\( (57 + 72) + (1 + 25) + 12 = 165 \\)" ]
[{"extraction_type": "final", "total_spans_found": 1, "non_overlapping_spans": 1, "span_start": 1700, "span_end": 1818, "original_span": "Step 1: 72 + 12 = 84 \nStep 2: 84 + 57 = 141 \nStep 3: 141 - 114 = 27 \nStep 4: 27 * 6 = 162 \nStep 5: 162 + 3 = 165"}, {"extraction_type": "final", "total_spans_found": 1, "non_...
[{"is_correct": false, "error": "Line uses unavailable numbers: Step 3: 141 - 114 = 27 ", "answer_block": "Step 1: 72 + 12 = 84 \nStep 2: 84 + 57 = 141 \nStep 3: 141 - 114 = 27 \nStep 4: 27 * 6 = 162 \nStep 5: 162 + 3 = 165", "final_answer": 27.0}, {"is_correct": false, "error": "Could not evaluate expression", "a...
[ [ false ], [ false ], [ true ], [ false ] ]
[ [ "Step 1: 72 + 12 = 84 \nStep 2: 84 + 57 = 141 \nStep 3: 141 - 114 = 27 \nStep 4: 27 * 6 = 162 \nStep 5: 162 + 3 = 165" ], [ "\\(165 = 12 + 25 + 1 + 72 - 57\\)" ], [ "((57 + 72) - 1) + 25 + 12 = 165" ], [ "\\( (57 + 72) + (1 + 25) + 12 = 165 \\)" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1700, "span_text": "Step 1: 72 + 12 = 84 \nStep 2: 84 + 57 = 141 \nStep 3: 141 - 114 = 27 \nStep 4: 27 * 6 = 162 \nStep 5: 162 + 3 = 165", "extraction_type": "internal", "span_start": 1700, "span_end": 1818, "original_span": "Step 1: 72 + 12 = 84 \nSt...
[[{"is_correct": false, "error": "Line uses unavailable numbers: Step 3: 141 - 114 = 27 ", "answer_block": "Step 1: 72 + 12 = 84 \nStep 2: 84 + 57 = 141 \nStep 3: 141 - 114 = 27 \nStep 4: 27 * 6 = 162 \nStep 5: 162 + 3 = 165", "final_answer": 27.0}], [{"is_correct": false, "error": "Could not evaluate expression",...
{ "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, ...
2025-11-10
test
main
TAUR-dev/M-AT_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860940
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
73*(18 - 4) - 54 - 69
countdown_5arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 259, \\\"numbers(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 259, \"numbers\": [69, 18, 54, 73, 4], \"targ(...TRUNCATED)
["First, I'll try to understand the problem better by writing out a plan and go really deep into det(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.10:10080/v1","backend":"litellm","chat_template_applied":true,"generatio(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ null, "(69 * 18) + 73", null, "(73 * 69) - 18 + 54 + 4 = 899" ]
"[{\"extraction_type\": \"final\"}, {\"extraction_type\": \"final\", \"total_spans_found\": 1, \"non(...TRUNCATED)
"[{\"evaluation_method\": \"response_processor\", \"reason\": \"no_extracted_answer\", \"extraction_(...TRUNCATED)
[ [], [ false ], [], [ false ] ]
[ [], [ "(69 * 18) + 73" ], [], [ "(73 * 69) - 18 + 54 + 4 = 899" ] ]
"[[], [{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 1946, \"span_text\": \"(...TRUNCATED)
"[[], [{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_bloc(...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_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860941
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
19*(79 - 71) + 62 + 87
countdown_5arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 260, \\\"numbers(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 260, \"numbers\": [87, 19, 71, 79, 62], \"tar(...TRUNCATED)
["Rewritten Question:\n\nCreate an equation using the numbers from the given set {87, 19, 71, 79, 62(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.10:10080/v1","backend":"litellm","chat_template_applied":true,"generatio(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
["((87 - 79) + 19)","No valid equation meets all requirements.","(87 - 79) + 71 + 62 = 301","87 - 79(...TRUNCATED)
"[{\"extraction_type\": \"final\", \"total_spans_found\": 1, \"non_overlapping_spans\": 1, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\": (...TRUNCATED)
[ [ false ], [ false ], [ false ], [ false ] ]
[["((87 - 79) + 19)"],["No valid equation meets all requirements."],["(87 - 79) + 71 + 62 = 301"],["(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 806, \"span_text\": \"((87 (...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_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860942
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-eval_sft
"Answer the following problem. Explain your reasoning step by step. When you are finished, give your(...TRUNCATED)
7*(97 + 7) - 38 - 34
countdown_5arg
countdown
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"source_dataset\\\": \\\"countdown\\\", \\\"source_index\\\": 262, \\\"numbers(...TRUNCATED)
test
"{\"source_dataset\": \"countdown\", \"source_index\": 262, \"numbers\": [97, 7, 7, 34, 38], \"targe(...TRUNCATED)
["Rewritten Question:\n\nCreate an equation using the numbers from the given list [97, 7, 7, 34, 38](...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.10:10080/v1","backend":"litellm","chat_template_applied":true,"generatio(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-AT_star_sft-sft", "n_responses": 4 }
[ false, false, false, false ]
[ "(97 + 7) * 7 - 34 + 38", "97 * 7 - 38 + 7 + 34 = 656", "97 + 7 + 7 + 34 - 38 + 7", "(97 * 7) - 34" ]
"[{\"extraction_type\": \"final\", \"total_spans_found\": 1, \"non_overlapping_spans\": 1, \"span_st(...TRUNCATED)
"[{\"is_correct\": false, \"error\": \"Expression does not equal stated result\", \"answer_block\": (...TRUNCATED)
[ [ false ], [ false ], [ false ], [ false ] ]
[["(97 + 7) * 7 - 34 + 38"],["97 * 7 - 38 + 7 + 34 = 656"],["97 + 7 + 7 + 34 - 38 + 7"],["(97 * 7) -(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 315, \"span_text\": \"(97 +(...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_star_sft-sft
0
eval_sft
1
2025-11-10T01:46:55.860944
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg-eval_sft
End of preview. Expand in Data Studio

Experiment Tracker: FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg

Experiment Description: Evaluation experiment for task countdown_5arg from FinEval_16k_fulleval_AT_STAR-SFT

Start Time: 2025-11-10T01:16:41.929294

Tracker Dataset: TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__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_STAR-SFT-countdown_5arg__v1', 'experiment_metadata')

# Load complete training datasets
sft_data = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'training_data__sft')
sft_metadata = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'training_data__sft_metadata')

# Load complete configurations
sft_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'hyperparameters__sft')
rl_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'hyperparameters__rl')

# Load stage-specific logs
sft_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'logs__sft')
rl_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'logs__rl')

# Load evaluation results with annotations
sft_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__v1', 'evals_eval_sft')
rl_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_AT_STAR-SFT-countdown_5arg__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_STAR-SFT-countdown_5arg - {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