Dataset Viewer
Auto-converted to Parquet Duplicate
question
string
answer
string
task_config
string
task_source
string
prompt
list
model_responses
list
model_responses__eval_is_correct
list
all_other_columns
string
original_split
string
metadata
string
model_responses__best_of_n_atags
list
model_responses__best_of_n_atags__finish_reason_length_flags
list
model_responses__best_of_n_atags__length_partial_responses
list
prompt__best_of_n_atags__metadata
dict
model_responses__best_of_n_atags__metadata
dict
model_responses__best_of_n_atags__eval_is_correct
list
model_responses__best_of_n_atags__eval_extracted_answers
list
model_responses__best_of_n_atags__eval_extraction_metadata
string
model_responses__best_of_n_atags__eval_evaluation_metadata
string
model_responses__best_of_n_atags__internal_answers__eval_is_correct
list
model_responses__best_of_n_atags__internal_answers__eval_extracted_answers
list
model_responses__best_of_n_atags__internal_answers__eval_extraction_metadata
string
model_responses__best_of_n_atags__internal_answers__eval_evaluation_metadata
string
model_responses__best_of_n_atags__metrics
dict
eval_date
string
split
string
revision_name
string
model_path
string
checkpoint_step
int64
stage_name
string
stage_number
int64
timestamp
string
eval_repo_id
string
30 x 62
1860
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n30 x 62\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 30, \"arg2\": 62, \"solution\": 1860}", "task_source": "longmult"}
test
{"arg1": 30, "arg2": 62, "solution": 1860}
[ "<think><sample>\nFirst, I need to multiply 30 and 62. That seems straightforward.\nI can use long multiplication or mental math. Let me do it carefully.\n\nWhen multiplying by 60 (since 62 has a zero at the end), 30 times 60 is 1800. But since we have an extra zero from 62, that doesn't change the total; it just m...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ true, false, true, true ]
[ "1860", "1920", "1860", "1860" ]
[{"extraction_type": "final", "total_spans_found": 5, "non_overlapping_spans": 5, "span_start": 3199, "span_end": 3205, "original_span": "1860"}, {"extraction_type": "final", "total_spans_found": 6, "non_overlapping_spans": 6, "span_start": 3741, "span_end": 3747, "original_span": "1920"}, {"extraction_type": "final", ...
[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ false, true, true, true, true ], [ false, true, true, true, true, false ], [ false, true, true, false, false, true, true ], [ false, true, true, true, true ] ]
[ [ "(your answer)", "1860", "1860", "1860", "1860" ], [ "(your answer)", "1860", "1860", "1860", "1860", "1920" ], [ "(your answer)", "1860", "1860", "(your answer)", "(your answer)", "1860", "1860" ], [ "(your answer)", "186...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 952, "span_text": "(your answer)", "extraction_type": "internal", "span_start": 952, "span_end": 967, "original_span": "(your answer)"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 1094, "span_text": "1860", "extraction_type": "internal",...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}], [{"evaluation_method": "legacy", "le...
{ "flips_by": [ 0, 1, 1, 0 ], "flips_total": 2, "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": [ 3, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077453
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
33 x 67
2211
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n33 x 67\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 33, \"arg2\": 67, \"solution\": 2211}", "task_source": "longmult"}
test
{"arg1": 33, "arg2": 67, "solution": 2211}
[ "<think><sample>\nFirst, I need to multiply these two numbers: 33 and 67.\nI can use long multiplication or break it down, but let me see if there's a pattern or shortcut that might make it easier.\n\nOne way is to add them directly since they're both three-digit numbers. But let me calculate that:\n\n30 * 67 = 201...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ true, true, false, true ]
[ "2211", "2211", "2409", "33 * 67 = 2211" ]
[{"extraction_type": "final", "total_spans_found": 11, "non_overlapping_spans": 11, "span_start": 4027, "span_end": 4033, "original_span": "2211"}, {"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 3496, "span_end": 3500, "original_span": "2211"}, {"extraction_type": "final"...
[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ false, false, false, false, false, false, true, false, false, true, true ], [ false, true, true, true ], [ false, false, false, false, false, false, false ], [ false, true, false, false, false, ...
[ [ "(your answer)", "(your answer)", "...", "...", "(your answer)", "...", "2211", "...", "594", "2211", "2211" ], [ "(your answer)", "2211", "2211", "2211" ], [ "2441", "(your answer)", "33 × 67 = ?", "(your answer)", "(your ans...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1975, "span_text": "(your answer)", "extraction_type": "internal", "span_start": 1975, "span_end": 1988, "original_span": "(your answer)"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 2927, "span_text": "(your answer)", "extraction_type":...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "...
{ "flips_by": [ 1, 0, 0, 1 ], "flips_total": 2, "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": [ 1, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077462
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
74 x 60
4440
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n74 x 60\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 74, \"arg2\": 60, \"solution\": 4440}", "task_source": "longmult"}
test
{"arg1": 74, "arg2": 60, "solution": 4440}
[ "<think><sample>\nFirst, I need to multiply 74 by 60. That seems straightforward, but let me calculate it.\n<compute>\n74 times 10 is 740 because 70 times 10 is 700 and 4 times 10 is 40, so 700+40=740.\n74 times 6 is 444 because 70 times 6 is 420 and 4 times 6 is 24, so 420+24=444.\nNow add them together: 740 + 444...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ true, true, true, true ]
[ "4440", "4440", "4440", "4440" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 3052, "span_end": 3056, "original_span": "4440"}, {"extraction_type": "final", "total_spans_found": 5, "non_overlapping_spans": 5, "span_start": 2773, "span_end": 2777, "original_span": "4440"}, {"extraction_type": "final", ...
[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ true, true ], [ false, false, true, true, true ], [ true, true, true, true, true ], [ false, false, true, true ] ]
[ [ "4440", "4440" ], [ "...", "(your answer)", "4440", "4440", "4440" ], [ "4440", "4440", "4440", "4440", "4440" ], [ "(your answer )", "(your answer )", "4440", "4440" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 2317, "span_text": "4440", "extraction_type": "internal", "span_start": 2317, "span_end": 2323, "original_span": "4440"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 3052, "span_text": "4440", "extraction_type": "internal", "span_start": ...
[[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}], [{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "l...
{ "flips_by": [ 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 4, "pass_at_n": 1, "percent_correct": 1, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 1, 2, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077464
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
69 x 45
3105
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n69 x 45\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 69, \"arg2\": 45, \"solution\": 3105}", "task_source": "longmult"}
test
{"arg1": 69, "arg2": 45, "solution": 3105}
[ "<think><sample>\nFirst, I need to multiply these two numbers: 69 and 45. The problem says \"When you are finished, give your answer in this format\", so I'll write it as an expression first.\n<answer>\n69 * 45\n</answer>\n</sample>\n<reflect>\nNow, for the actual computation, I should calculate that. But let me se...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ true, false, true, true ]
[ "3105", "3015", "3105", "3105" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 2989, "span_end": 2995, "original_span": "3105"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 3679, "span_end": 3685, "original_span": "3015"}, {"extraction_type": "final", ...
[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ false, true ], [ false, false, false ], [ true, false, true, true ], [ false, true, true ] ]
[ [ "69 * 45", "3105" ], [ "(your answer )", "42", "3015" ], [ "3105", "(your answer)", "3105", "3105" ], [ "(your answer )", "3105", "3105" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 196, "span_text": "69 * 45", "extraction_type": "internal", "span_start": 196, "span_end": 205, "original_span": "69 * 45"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 2989, "span_text": "3105", "extraction_type": "internal", "span_start...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}], [{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}], [{"evaluation_method": "legacy"...
{ "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": [ 0, 0, 0, 0 ], "reflect_close": [ 0, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077465
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
73 x 71
5183
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n73 x 71\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 73, \"arg2\": 71, \"solution\": 5183}", "task_source": "longmult"}
test
{"arg1": 73, "arg2": 71, "solution": 5183}
[ "<think><sample>\nFirst, I need to multiply these two numbers: 73 and 71.\nI can do this directly or use the distributive property, but since they're both close to 70, I might as well add them directly.\n\nSo, 73 + 71 = ?\n\nTo calculate that:\n\n- 73 + 70 = 143 (because 70 + 70 is 140, plus 3 more gives 143)\n\nNo...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ false, false, true, false ]
[ "5230", "5254", "5183", "516110" ]
[{"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 5826, "span_end": 5832, "original_span": "5230"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 3813, "span_end": 3819, "original_span": "5254"}, {"extraction_type": "final", ...
[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}]
[ [ false, false, false ], [ false, false, false ], [ true, true ], [ false, false ] ]
[ [ "(your answer)", "5230", "5230" ], [ "25", "5254", "5254" ], [ "5183", "5183" ], [ "5113", "516110" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 757, "span_text": "(your answer)", "extraction_type": "internal", "span_start": 757, "span_end": 772, "original_span": "(your answer)"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 3771, "span_text": "5230", "extraction_type": "internal",...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}], [{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy",...
{ "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": [ 1, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077467
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
59 x 36
2124
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n59 x 36\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 59, \"arg2\": 36, \"solution\": 2124}", "task_source": "longmult"}
test
{"arg1": 59, "arg2": 36, "solution": 2124}
[ "<think><sample>\nFirst, I need to multiply these two numbers: 59 and 36.\nThe problem says \"when you are finished\", so let me calculate that.\nI'll use long multiplication or a calculator for accuracy.\nBut since it's simple, I can do it mentally or with paper.\n\nLet me break it down:\n\n- 36 times 50 is 1800 (...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ true, true, true, true ]
[ "2124", "2124", "2124", "2124" ]
[{"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 3421, "span_end": 3427, "original_span": "2124"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 4248, "span_end": 4254, "original_span": "2124"}, {"extraction_type": "final", ...
[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ true, true, true ], [ true, true, true ], [ true, false, true, true ], [ false, true, true, true ] ]
[ [ "2124", "2124", "2124" ], [ "2124", "2124", "2124" ], [ "2124", "(your answer)", "2124", "2124" ], [ "(your answer)", "2124", "2124", "2124" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 957, "span_text": "2124", "extraction_type": "internal", "span_start": 957, "span_end": 963, "original_span": "2124"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 3045, "span_text": "2124", "extraction_type": "internal", "span_start": 304...
[[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}], [{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "leg...
{ "flips_by": [ 0, 0, 1, 0 ], "flips_total": 1, "num_correct": 4, "pass_at_n": 1, "percent_correct": 1, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 2, 2, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077468
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
23 x 24
552
longmult_2dig
longmult
[ { "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\nWhat is the solution to the multiplication equation below?\n23 x 24\n\nGive your answer in the following format:\n<answer>\n(your...
[]
[]
{"metadata": "{\"arg1\": 23, \"arg2\": 24, \"solution\": 552}", "task_source": "longmult"}
test
{"arg1": 23, "arg2": 24, "solution": 552}
[ "<think><sample>\nFirst, I need to multiply 23 and 24. Let me do that calculation.\nI can break it down or use the standard method. Since both numbers are close to 20, I might use that for simplicity.\n\nMultiplying by 25 is easy: 25 × 24 = (100 - 75) × 24 = 100×24 - 75×24 = 2400 - 1800 = 600.\n\nBut since we're mu...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://127.0.0.1:40881/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_vllm...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft", "n_responses": 4 }
[ true, true, true, true ]
[ "552", "552", "552", "552" ]
[{"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 2506, "span_end": 2509, "original_span": "552"}, {"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 2200, "span_end": 2205, "original_span": "552"}, {"extraction_type": "final", "t...
[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ true, false, true, true ], [ true, true, true, true ], [ true, false, true, true, true, true ], [ false, true, true ] ]
[ [ "552", "(your answer)", "552", "552" ], [ "552", "552", "552", "552" ], [ "552", "(your answer)", "552", "552", "552", "552" ], [ "(your answer )", "552", "552" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 615, "span_text": "552", "extraction_type": "internal", "span_start": 615, "span_end": 620, "original_span": "552"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 736, "span_text": "(your answer)", "extraction_type": "internal", "span_start...
[[{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}], [{"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "le...
{ "flips_by": [ 1, 0, 1, 0 ], "flips_total": 2, "num_correct": 4, "pass_at_n": 1, "percent_correct": 1, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 2, 3, ...
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077470
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
47 x 41
1927
longmult_2dig
longmult
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"arg1\\\": 47, \\\"arg2\\\": 41, \\\"solution\\\": 1927}\", \"task_source\": \(...TRUNCATED)
test
{"arg1": 47, "arg2": 41, "solution": 1927}
["<think><sample>\nFirst, I need to multiply 47 and 41. That seems like a straightforward multiplica(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://127.0.0.1:40881/v1","backend":"litellm","chat_template_applied":true,"generation_(...TRUNCATED)
{"backend":"litellm","model_name":"hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft","n_responses":(...TRUNCATED)
[ false, false, true, true ]
[ null, "1967", "1927", "1927" ]
"[{\"extraction_type\": \"final\"}, {\"extraction_type\": \"final\", \"total_spans_found\": 3, \"non(...TRUNCATED)
"[{\"evaluation_method\": \"response_processor\", \"reason\": \"no_extracted_answer\", \"extraction_(...TRUNCATED)
[ [], [ false, true, false ], [ false, false, true, true ], [ false, true, true, false, true ] ]
[[],["(47 * 41) = ?","1927","1967"],["(your answer )","(your answer)","1927","1927"],["(47 * 41) = ?(...TRUNCATED)
"[[], [{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 80, \"span_text\": \"(4(...TRUNCATED)
"[[], [{\"evaluation_method\": \"legacy\", \"legacy_result\": false}, {\"evaluation_method\": \"lega(...TRUNCATED)
{"flips_by":[0,1,0,1],"flips_total":2,"num_correct":2,"pass_at_n":1,"percent_correct":0.5,"skill_cou(...TRUNCATED)
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077471
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
89 x 46
4094
longmult_2dig
longmult
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"arg1\\\": 89, \\\"arg2\\\": 46, \\\"solution\\\": 4094}\", \"task_source\": \(...TRUNCATED)
test
{"arg1": 89, "arg2": 46, "solution": 4094}
["<think><sample>\nFirst, I need to multiply these two numbers: 89 and 46.\nI can break it down or u(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://127.0.0.1:40881/v1","backend":"litellm","chat_template_applied":true,"generation_(...TRUNCATED)
{"backend":"litellm","model_name":"hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft","n_responses":(...TRUNCATED)
[ true, true, true, true ]
[ "4094", "4094", "4094", "4094" ]
"[{\"extraction_type\": \"final\", \"total_spans_found\": 4, \"non_overlapping_spans\": 4, \"span_st(...TRUNCATED)
"[{\"evaluation_method\": \"legacy\", \"legacy_result\": true}, {\"evaluation_method\": \"legacy\", (...TRUNCATED)
[[false,false,true,true],[true,false,false,false,true,true],[true,true],[true,false,true,true,true,t(...TRUNCATED)
[["(your answer)","(your answer)","4094","4094"],["4094","(your answer )","(your answer )","(your an(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 1118, \"span_text\": \"(you(...TRUNCATED)
"[[{\"evaluation_method\": \"legacy\", \"legacy_result\": false}, {\"evaluation_method\": \"legacy\"(...TRUNCATED)
{"flips_by":[0,1,0,1],"flips_total":2,"num_correct":4,"pass_at_n":1,"percent_correct":1.0,"skill_cou(...TRUNCATED)
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077472
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
30 x 35
1050
longmult_2dig
longmult
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"metadata\": \"{\\\"arg1\\\": 30, \\\"arg2\\\": 35, \\\"solution\\\": 1050}\", \"task_source\": \(...TRUNCATED)
test
{"arg1": 30, "arg2": 35, "solution": 1050}
["<think><sample>\nFirst, I need to multiply 30 and 35. That seems straightforward.\n<action>\nMulti(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://127.0.0.1:40881/v1","backend":"litellm","chat_template_applied":true,"generation_(...TRUNCATED)
{"backend":"litellm","model_name":"hosted_vllm/TAUR-dev/M-r1_translated_BASELINE-sft","n_responses":(...TRUNCATED)
[ true, true, true, true ]
[ "1050", "1050", "1050", "1050" ]
"[{\"extraction_type\": \"final\", \"total_spans_found\": 3, \"non_overlapping_spans\": 3, \"span_st(...TRUNCATED)
"[{\"evaluation_method\": \"legacy\", \"legacy_result\": true}, {\"evaluation_method\": \"legacy\", (...TRUNCATED)
[ [ false, false, true ], [ true, false, false, true, true ], [ false, false, true, true ], [ true ] ]
[["(your answer)","24","1050"],["1050","(your answer)","(your answer )","1050","1050"],["(your answe(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 482, \"span_text\": \"(your(...TRUNCATED)
"[[{\"evaluation_method\": \"legacy\", \"legacy_result\": false}, {\"evaluation_method\": \"legacy\"(...TRUNCATED)
{"flips_by":[0,1,0,0],"flips_total":1,"num_correct":4,"pass_at_n":1,"percent_correct":1.0,"skill_cou(...TRUNCATED)
2025-11-02
test
main
TAUR-dev/M-r1_translated_BASELINE-sft
0
eval_sft
1
2025-11-02T01:32:18.077474
TAUR-dev/D-EVAL__standard_eval_v3__Fin16kEval_SFT_r1_translated-longmult_2dig-eval_sft
End of preview. Expand in Data Studio

Experiment Tracker: Fin16kEval_SFT_r1_translated-longmult_2dig

Experiment Description: Evaluation experiment for task longmult_2dig from Fin16kEval_SFT_r1_translated

Start Time: 2025-11-02T00:49:08.449702

Tracker Dataset: TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__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__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'experiment_metadata')

# Load complete training datasets
sft_data = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'training_data__sft')
sft_metadata = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'training_data__sft_metadata')

# Load complete configurations
sft_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'hyperparameters__sft')
rl_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'hyperparameters__rl')

# Load stage-specific logs
sft_logs = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'logs__sft')
rl_logs = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'logs__rl')

# Load evaluation results with annotations
sft_eval_results = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__v1', 'evals_eval_sft')
rl_eval_results = load_dataset('TAUR-dev/D-ExpTracker__Fin16kEval_SFT_r1_translated-longmult_2dig__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 - Fin16kEval_SFT_r1_translated-longmult_2dig - {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