Dataset Viewer
Auto-converted to Parquet Duplicate
question
stringlengths
21
279
answer
stringlengths
3
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
200
264
original_split
stringclasses
1 value
answer_index
int64
0
4
answer_key
stringclasses
5 values
choices
dict
id
stringlengths
32
34
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
560
987
model_responses__best_of_n_atags__eval_evaluation_metadata
stringclasses
16 values
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
1.07k
37.8k
model_responses__best_of_n_atags__internal_answers__eval_evaluation_metadata
stringlengths
350
9.58k
model_responses__best_of_n_atags__metrics
dict
eval_date
stringdate
2025-11-09 00:00:00
2025-11-09 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
2025-11-09 18:07:52
2025-11-09 18:07:52
eval_repo_id
stringclasses
1 value
Where would you keep a rug near your front door?
living room
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nWhere would you keep a rug near your front door?\n\n# Answer Choices\nA - persia\nB - desk\nC - table\nD - living room\nE - hal...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["persia", "desk", "table", "living room", "hall"]}, "answer_key": "D", "answer_index": 3, "task_source": "csqa", "id": "22015315e7ff79386877828b4fa27799"}
test
3
D
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "persia", "desk", "table", "living room", "hall" ] }
22015315e7ff79386877828b4fa27799
[ "<think>\n<sample>\nTo determine where to keep a rug near your front door, let's consider each option:\n\nA. Persia: This refers to Iran and has no direct relevance to keeping rugs near doors.\nB. Desk: A rug on a desk might clutter space but doesn't logically fit near a front door.\nC. Table: Similar to a desk, pl...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ false, true, true, false ]
[ "E", "D", "D", "E" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 1932, "span_end": 1935, "original_span": "E"}, {"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 799, "span_end": 802, "original_span": "D"}, {"extraction_type": "final", "total_s...
[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}]
[ [ false, false ], [ true, true ], [ true, true, true ], [ false, false ] ]
[ [ "E", "E" ], [ "D", "D" ], [ "D", "D", "D" ], [ "E", "E" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1850, "span_text": "E", "extraction_type": "internal", "span_start": 1850, "span_end": 1853, "original_span": "E"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 1932, "span_text": "E", "extraction_type": "internal", "span_start": 1932, "sp...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"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", ...
{ "flips_by": [ 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 2, "pass_at_n": 1, "percent_correct": 0.5, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 0, 0, 0, 0 ], "reflect_close": [ 0, 0...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562951
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
Danny found an old film in a sealed what?
cabinet
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nDanny found an old film in a sealed what?\n\n# Answer Choices\nA - clingfilm\nB - disneyland\nC - cave\nD - cabinet\nE - movie\...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["clingfilm", "disneyland", "cave", "cabinet", "movie"]}, "answer_key": "D", "answer_index": 3, "task_source": "csqa", "id": "a328285c6212c899e335c45db3c49ffd"}
test
3
D
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "clingfilm", "disneyland", "cave", "cabinet", "movie" ] }
a328285c6212c899e335c45db3c49ffd
[ "<think>\n<sample>\nTo solve this problem, let's analyze each option:\n\nA. Clingfilm - This is commonly used for sealing food items but not typically associated with finding films.\nB. Disneyland - While fun and enjoyable, it's not something one would find an old film in.\nC. Cave - Caves can have preserved artifa...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ false, true, false, false ]
[ "C", "D", "C", "C" ]
[{"extraction_type": "final", "total_spans_found": 4, "non_overlapping_spans": 4, "span_start": 2657, "span_end": 2660, "original_span": "C"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 2184, "span_end": 2187, "original_span": "D"}, {"extraction_type": "final", "total...
[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}]
[ [ false, false, false, false ], [ true, true, true ], [ false, false, false ], [ false, false, false ] ]
[ [ "C", "C", "C", "C" ], [ "D", "D", "D" ], [ "C", "C", "C" ], [ "C", "C", "C" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 794, "span_text": "C", "extraction_type": "internal", "span_start": 794, "span_end": 797, "original_span": "C"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 1635, "span_text": "C", "extraction_type": "internal", "span_start": 1635, "span_...
[[{"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": true}, {"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": [ 1, 0, 1, 0 ], "reflect_close": [ 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562968
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
The earth is one planet in what?
solar system
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nThe earth is one planet in what?\n\n# Answer Choices\nA - tree\nB - orbit\nC - solar system\nD - fotograph\nE - dreams\n\nGive ...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["tree", "orbit", "solar system", "fotograph", "dreams"]}, "answer_key": "C", "answer_index": 2, "task_source": "csqa", "id": "17eafc807b198236faf06a66f4c05313"}
test
2
C
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "tree", "orbit", "solar system", "fotograph", "dreams" ] }
17eafc807b198236faf06a66f4c05313
[ "<think>\n<sample>\nTo solve this problem, let's think through it logically:\n\n1. The question asks about the Earth being part of something.\n2. We need to identify which option correctly describes where the Earth fits into the universe.\n\nNow, let's analyze each option:\n\nA. Tree - This does not relate to plane...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ true, true, true, true ]
[ "C", "C", "C", "C" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 826, "span_end": 829, "original_span": "C"}, {"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 1066, "span_end": 1069, "original_span": "C"}, {"extraction_type": "final", "total_s...
[{"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 ], [ true, true ] ]
[ [ "C", "C" ], [ "C", "C" ], [ "C", "C", "C" ], [ "C", "C" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 744, "span_text": "C", "extraction_type": "internal", "span_start": 744, "span_end": 747, "original_span": "C"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 826, "span_text": "C", "extraction_type": "internal", "span_start": 826, "span_en...
[[{"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", "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": [ 0, 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562970
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
What is someone operating a vehicle likely to be accused of after becoming inebriated?
drunk driving
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nWhat is someone operating a vehicle likely to be accused of after becoming inebriated?\n\n# Answer Choices\nA - punish\nB - arr...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["punish", "arrest", "automobile accidents", "drunk driving", "talking nonsense"]}, "answer_key": "D", "answer_index": 3, "task_source": "csqa", "id": "cfa980561efe82e7ae7080d4f081b463"}
test
3
D
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "punish", "arrest", "automobile accidents", "drunk driving", "talking nonsense" ] }
cfa980561efe82e7ae7080d4f081b463
[ "<think>\n<sample>\nTo solve this problem, let's consider each option carefully:\n\nA. Punish - This refers to legal action taken against someone who commits a crime but does not directly relate to the act of being intoxicated and driving.\nB. Arrest - Similar to punishment, arresting someone involves legal action ...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ true, true, true, true ]
[ "D", "D", "D", "D" ]
[{"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 2080, "span_end": 2081, "original_span": "D"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 2004, "span_end": 2007, "original_span": "D"}, {"extraction_type": "final", "total...
[{"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, true, true ], [ true, true, true, true, true ] ]
[ [ "D", "D", "D" ], [ "D", "D", "D" ], [ "D", "D", "D" ], [ "D", "D", "D", "D", "D" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 965, "span_text": "D", "extraction_type": "internal", "span_start": 965, "span_end": 968, "original_span": "D"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 2001, "span_text": "D", "extraction_type": "internal", "span_start": 2001, "span_...
[[{"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, 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": [ 1, 0, 0, 1 ], "reflect_close": [ 0, 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562972
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
The town house went right to the curb, a slot effectively made a mailbox of the what?
front door
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nThe town house went right to the curb, a slot effectively made a mailbox of the what?\n\n# Answer Choices\nA - apartment buildi...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["apartment building", "front door", "back door", "street corner", "porch"]}, "answer_key": "B", "answer_index": 1, "task_source": "csqa", "id": "70d3ebc00b165d9d08f9491a1dd85034"}
test
1
B
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "apartment building", "front door", "back door", "street corner", "porch" ] }
70d3ebc00b165d9d08f9491a1dd85034
[ "<think>\n<sample>\nTo solve this problem, let's analyze each option:\n\nA. Apartment building - An apartment building typically has multiple units on one floor and does not have slots for mailboxes.\nB. Front door - The front door is where people enter their homes but doesn't usually function as a mailbox.\nC. Bac...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ false, false, false, false ]
[ "E", "front door", "E", "porch" ]
[{"extraction_type": "final", "total_spans_found": 8, "non_overlapping_spans": 8, "span_start": 11524, "span_end": 11527, "original_span": "E"}, {"extraction_type": "final", "total_spans_found": 7, "non_overlapping_spans": 7, "span_start": 7986, "span_end": 7998, "original_span": "front door"}, {"extraction_type": "fin...
[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}]
[ [ false, false, false, true, false, false, false, false ], [ false, false, false, false, false, false, false ], [ false, false, false, false, false, false, false, false, false, false, false ], [ f...
[ [ "porch", "street corner", "porch", "B", "e", "porch", "E", "E" ], [ "D", "E", "E", "curb", "No valid answer found among the options provided.", "front door", "front door" ], [ "E", "E", "E", "E", "E", "E", "E", "E"...
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 1216, "span_text": "porch", "extraction_type": "internal", "span_start": 1216, "span_end": 1223, "original_span": "porch"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 2833, "span_text": "street corner", "extraction_type": "internal", "sp...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "l...
{ "flips_by": [ 1, 0, 0, 0 ], "flips_total": 1, "num_correct": 0, "pass_at_n": 0, "percent_correct": 0, "skill_count": { "answer_revision": [ 0, 0, 0, 0 ], "best_of_n": [ 2, 1, 0, 1 ], "reflect_close": [ 0, 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562974
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
Where is a salt shaker most often kept?
table setting
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nWhere is a salt shaker most often kept?\n\n# Answer Choices\nA - cruet\nB - table setting\nC - kitchen cupboard\nD - cabinet\nE...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["cruet", "table setting", "kitchen cupboard", "cabinet", "store"]}, "answer_key": "B", "answer_index": 1, "task_source": "csqa", "id": "c2a8c6814ed3e207771cfc23b3b42cf1"}
test
1
B
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "cruet", "table setting", "kitchen cupboard", "cabinet", "store" ] }
c2a8c6814ed3e207771cfc23b3b42cf1
[ "<think>\n<sample>\nTo determine where a salt shaker is most often kept, let's consider each option:\n\nA. Cruet: A cruet typically holds multiple condiments together, but it does not specifically mean it keeps a salt shaker inside.\nB. Table setting: While a salt shaker might be part of a table setting, it doesn't...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ false, false, false, true ]
[ "C", "C", "C", "B" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 993, "span_end": 996, "original_span": "C"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 2044, "span_end": 2047, "original_span": "C"}, {"extraction_type": "final", "total_s...
[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ false, false ], [ false, false, false ], [ false, false, false, false ], [ true, true ] ]
[ [ "C", "C" ], [ "C", "C", "C" ], [ "C", "C", "C", "C" ], [ "B", "B" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 911, "span_text": "C", "extraction_type": "internal", "span_start": 911, "span_end": 914, "original_span": "C"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 993, "span_text": "C", "extraction_type": "internal", "span_start": 993, "span_en...
[[{"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, 1, 0, 0 ], "reflect_close": [ 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562976
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
Everybody was changing into costumes in the dressing room, it was almost time to take the what stage?
theater
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nEverybody was changing into costumes in the dressing room, it was almost time to take the what stage?\n\n# Answer Choices\nA - ...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["theater", "train", "bathhouse", "dwelling", "actors and actresses"]}, "answer_key": "A", "answer_index": 0, "task_source": "csqa", "id": "db63bf66a8bfd16e5103cbdd350f5202"}
test
0
A
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "theater", "train", "bathhouse", "dwelling", "actors and actresses" ] }
db63bf66a8bfd16e5103cbdd350f5202
[ "<think>\n<sample>\nTo solve this problem, let's analyze each option:\n\nA. Theater - This is a common place where performances happen, but not necessarily for taking roles.\nB. Train - This doesn't make sense as a location for preparing costumes before going on stage.\nC. Bathhouse - While people might change clot...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ false, true, true, true ]
[ "theater", "A", "A", "A" ]
[{"extraction_type": "final", "total_spans_found": 8, "non_overlapping_spans": 8, "span_start": 9953, "span_end": 9962, "original_span": "theater"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 1787, "span_end": 1790, "original_span": "A"}, {"extraction_type": "final", ...
[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}]
[ [ false, true, true, false, true, false, false, false ], [ true, true, true ], [ true, true, true ], [ true, true, true ] ]
[ [ "None of the above", "A", "A", "None of the above", "A", "E", "theater", "theater" ], [ "A", "A", "A" ], [ "A", "A", "A" ], [ "A", "A", "A" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 789, "span_text": "None of the above", "extraction_type": "internal", "span_start": 789, "span_end": 808, "original_span": "None of the above"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 2394, "span_text": "A", "extraction_type": "inter...
[[{"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "legacy_result": false}, {"evaluation_method": "legacy", "legacy_result": true}, {"evaluation_method": "legacy", "leg...
{ "flips_by": [ 2, 0, 0, 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": [ 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562978
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
What would someone use a personal key for?
own home
commonsenseQA
csqa
[ { "content": "Answer the following problem. Explain your reasoning step by step. When you are finished, give your answer in this format: <answer>(letter choice)</answer>.\n\n# Problem\nWhat would someone use a personal key for?\n\n# Answer Choices\nA - car stand\nB - at hotel\nC - own home\nD - front door\nE - ...
[]
[]
{"choices": {"label": ["A", "B", "C", "D", "E"], "text": ["car stand", "at hotel", "own home", "front door", "bus depot"]}, "answer_key": "C", "answer_index": 2, "task_source": "csqa", "id": "1b3d286458a7e7f069222de0376d06da"}
test
2
C
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "car stand", "at hotel", "own home", "front door", "bus depot" ] }
1b3d286458a7e7f069222de0376d06da
[ "<think>\n<sample>\nTo determine what a personal key would be used for, let's consider each option:\n\nA. Car stand: A car stand is typically found in garages or parking lots and not related to personal keys.\nB. At hotel: While some hotels may provide guests with access cards, they don't usually have personal keys...
[ false, false, false, false ]
[ "", "", "", "" ]
{ "api_url": "http://10.32.37.22:9080/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-ablations__rl_ab_no_reflects-rl", "n_responses": 4 }
[ true, true, true, true ]
[ "C", "C", "C", "C" ]
[{"extraction_type": "final", "total_spans_found": 2, "non_overlapping_spans": 2, "span_start": 869, "span_end": 872, "original_span": "C"}, {"extraction_type": "final", "total_spans_found": 3, "non_overlapping_spans": 3, "span_start": 1722, "span_end": 1725, "original_span": "C"}, {"extraction_type": "final", "total_s...
[{"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, true ], [ true, true, true ] ]
[ [ "C", "C" ], [ "C", "C", "C" ], [ "C", "C", "C" ], [ "C", "C", "C" ] ]
[[{"pattern_used": "answer tag", "confidence": 1.0, "position": 787, "span_text": "C", "extraction_type": "internal", "span_start": 787, "span_end": 790, "original_span": "C"}, {"pattern_used": "answer tag", "confidence": 1.0, "position": 869, "span_text": "C", "extraction_type": "internal", "span_start": 869, "span_en...
[[{"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", "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": [ 0, 0, ...
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562980
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
What does playing soccer and winning lead to?
excitement
commonsenseQA
csqa
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"choices\": {\"label\": [\"A\", \"B\", \"C\", \"D\", \"E\"], \"text\": [\"excitement\", \"getting(...TRUNCATED)
test
0
A
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "excitement", "getting tired", "overtime", "anger", "fights" ] }
64ab884bd870f6f68146636b4cce921c
["<think>\n<sample>\nTo determine the correct answer, let's consider each option provided:\n\nA. Exc(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.22:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{"backend":"litellm","model_name":"hosted_vllm/TAUR-dev/M-ablations__rl_ab_no_reflects-rl","n_respon(...TRUNCATED)
[ true, false, false, true ]
[ "A", "excitement", "goals scored", "A" ]
"[{\"extraction_type\": \"final\", \"total_spans_found\": 2, \"non_overlapping_spans\": 2, \"span_st(...TRUNCATED)
"[{\"evaluation_method\": \"legacy\", \"legacy_result\": true}, {\"evaluation_method\": \"legacy\", (...TRUNCATED)
[ [ true, true ], [ false, false, false ], [ false, false, false, false, false, false ], [ true, true, true ] ]
[["A","A"],["D","excitement","excitement"],["D","None of the above options accurately describe what (...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 898, \"span_text\": \"A\", (...TRUNCATED)
"[[{\"evaluation_method\": \"legacy\", \"legacy_result\": true}, {\"evaluation_method\": \"legacy\",(...TRUNCATED)
{"flips_by":[0,0,0,0],"flips_total":0,"num_correct":2,"pass_at_n":1,"percent_correct":0.5,"skill_cou(...TRUNCATED)
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562982
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
"During the winter hunt he could hear every motion in the woods, this was because of the what of eve(...TRUNCATED)
stillness
commonsenseQA
csqa
[{"content":"Answer the following problem. Explain your reasoning step by step. When you are finishe(...TRUNCATED)
[]
[]
"{\"choices\": {\"label\": [\"A\", \"B\", \"C\", \"D\", \"E\"], \"text\": [\"being still\", \"silenc(...TRUNCATED)
test
3
D
{"label":["A","B","C","D","E"],"text":["being still","silence","stationary","stillness","standing st(...TRUNCATED)
8b2af2d865b7dc500427786c846eacaf
["<think>\n<sample>\nTo solve this problem, let's analyze each option:\n\nA. Being still - This mean(...TRUNCATED)
[ false, false, false, false ]
[ "", "", "", "" ]
{"api_url":"http://10.32.37.22:9080/v1","backend":"litellm","chat_template_applied":true,"generation(...TRUNCATED)
{"backend":"litellm","model_name":"hosted_vllm/TAUR-dev/M-ablations__rl_ab_no_reflects-rl","n_respon(...TRUNCATED)
[ false, false, true, false ]
[ "E", "E", "D", "E" ]
"[{\"extraction_type\": \"final\", \"total_spans_found\": 4, \"non_overlapping_spans\": 4, \"span_st(...TRUNCATED)
"[{\"evaluation_method\": \"legacy\", \"legacy_result\": false}, {\"evaluation_method\": \"legacy\",(...TRUNCATED)
[ [ true, false, false, false ], [ false, false ], [ false, true, true ], [ false, false, false ] ]
[["D","None of the above (as none of the options correctly describe the scenario)","E","E"],["E","E"(...TRUNCATED)
"[[{\"pattern_used\": \"answer tag\", \"confidence\": 1.0, \"position\": 1692, \"span_text\": \"D\",(...TRUNCATED)
"[[{\"evaluation_method\": \"legacy\", \"legacy_result\": true}, {\"evaluation_method\": \"legacy\",(...TRUNCATED)
{"flips_by":[1,0,0,0],"flips_total":1,"num_correct":1,"pass_at_n":1,"percent_correct":0.25,"skill_co(...TRUNCATED)
2025-11-09
test
main
TAUR-dev/M-ablations__rl_ab_no_reflects-rl
0
eval_rl
1
2025-11-09T18:07:52.562984
TAUR-dev/D-EVAL__standard_eval_v3__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA-eval_rl
End of preview. Expand in Data Studio

Experiment Tracker: FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA

Experiment Description: Evaluation experiment for task commonsenseQA from FinEval_16k_fulleval_3args_NoReflects-RL

Start Time: 2025-11-09T17:08:08.637372

Tracker Dataset: TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__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_3args_NoReflects-RL-commonsenseQA__v1', 'experiment_metadata')

# Load complete training datasets
sft_data = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'training_data__sft')
sft_metadata = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'training_data__sft_metadata')

# Load complete configurations
sft_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'hyperparameters__sft')
rl_hyperparams = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'hyperparameters__rl')

# Load stage-specific logs
sft_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'logs__sft')
rl_logs = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'logs__rl')

# Load evaluation results with annotations
sft_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__v1', 'evals_eval_sft')
rl_eval_results = load_dataset('TAUR-dev/D-ExpTracker__FinEval_16k_fulleval_3args_NoReflects-RL-commonsenseQA__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_3args_NoReflects-RL-commonsenseQA - {stage_name} - {SFT/RL}


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

Downloads last month
7