Dataset Viewer
Auto-converted to Parquet Duplicate
question
stringlengths
19
270
answer
stringlengths
3
28
task_config
stringclasses
1 value
task_source
stringclasses
1 value
prompt
listlengths
2
2
model_responses
listlengths
0
0
model_responses__eval_is_correct
listlengths
0
0
prompt_variant
stringclasses
1 value
choices
dict
answer_key
stringclasses
5 values
answer_index
int64
0
4
id
stringlengths
32
34
model_responses__mutated_prompts
listlengths
16
16
model_responses__mutated_prompts__finish_reason_length_flags
listlengths
16
16
model_responses__mutated_prompts__length_partial_responses
listlengths
16
16
prompt__mutated_prompts__metadata
dict
model_responses__mutated_prompts__metadata
dict
model_responses__mutated_prompts__eval_is_correct
listlengths
16
16
model_responses__mutated_prompts__eval_extracted_answers
listlengths
16
16
model_responses__mutated_prompts__eval_extraction_metadata
listlengths
16
16
model_responses__mutated_prompts__eval_evaluation_metadata
listlengths
16
16
model_responses__mutated_prompts__internal_answers__eval_is_correct
listlengths
16
16
model_responses__mutated_prompts__internal_answers__eval_extracted_answers
listlengths
16
16
model_responses__mutated_prompts__internal_answers__eval_extraction_metadata
listlengths
16
16
model_responses__mutated_prompts__internal_answers__eval_evaluation_metadata
listlengths
16
16
model_responses__mutated_prompts__metrics
dict
What do kids have to do before they can go outside a house?
open door
commonsenseQA
csqa
[ { "content": "You like to solve problems by intentionally making mistakes then fixing them, then giving an answer. Think step by step, introduce an error, recognize it, fix it, finish out the rest of the reasoning, then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, y...
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "distracting", "open door", "wonder about", "take shoes off", "become adults" ] }
B
1
c40bd55eec7c264c0fcfa929d4ec8fe9
[ "When you're ready, please provide your answer.", "First, let's consider the options given:\n\n1. **Distractions**: Kids might be distracted while inside a house, but going outside usually doesn't require distractions.\n2. **Opening a door**: This action would allow kids to leave the house, so opening a door seem...
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{ "api_url": "http://127.0.0.1:41177", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 4096, "n": 16, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vllm/Qw...
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, true, false, true, false, false, true, true, true, true, true, false, false, false, false, false ]
[ null, "B", "A", "B", null, "D", "B", "B", "B", "B", "B", "D", "E", "open door", "Supervision", "D" ]
[ { "extraction_type": "final", "non_overlapping_spans": null, "original_span": null, "span_end": null, "span_start": null, "total_spans_found": null }, { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "B", "span_end": 1266, "span_start": 1265,...
[ { "evaluation_method": "response_processor", "extraction_result": "{'extraction_type': 'final'}", "legacy_result": null, "reason": "no_extracted_answer" }, { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": true, "reason": null }, { "evaluation_m...
[ [], [ true ], [ false ], [ true ], [], [ false ], [ true ], [ true ], [ true ], [ true ], [ true ], [ false ], [ false ], [ false ], [ false ], [ false ] ]
[ [], [ "B" ], [ "A" ], [ "B" ], [], [ "D" ], [ "B" ], [ "B" ], [ "B" ], [ "B" ], [ "B" ], [ "D" ], [ "E" ], [ "open door" ], [ "Supervision" ], [ "D" ] ]
[ [], [ { "confidence": 1, "extraction_type": "internal", "original_span": "B", "pattern_used": "answer tag", "position": 1265, "span_end": 1266, "span_start": 1265, "span_text": "B" } ], [ { "confidence": 1, "extraction_type": "internal", ...
[ [], [ { "evaluation_method": "legacy", "legacy_result": true } ], [ { "evaluation_method": "legacy", "legacy_result": false } ], [ { "evaluation_method": "legacy", "legacy_result": true } ], [], [ { "evaluation_method": "legacy", ...
{ "flips_by": [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 7, "pass_at_n": 1, "percent_correct": 0.4375, "skill_count": { "answer_revision": [ 0, 0, 0, 0, 0, 0, ...
If air is held down by gravity, what does it usually cover.
surface of earth
commonsenseQA
csqa
[ { "content": "You like to solve problems by intentionally making mistakes then fixing them, then giving an answer. Think step by step, introduce an error, recognize it, fix it, finish out the rest of the reasoning, then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, y...
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "sky", "space shuttle", "cupboard", "surface of earth", "house" ] }
D
3
b025f44034cfec59c1b72c7675baf084
[ "The correct answer would be D - surface of earth.\n\nExplanation:\nWhen we consider how air behaves under normal conditions, we often think about its role in providing us with breathable air. However, when air is held down by gravity, it typically covers the entire surface of our planet, which is the Earth's atmos...
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{ "api_url": "http://127.0.0.1:41177", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 4096, "n": 16, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vllm/Qw...
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, true, true, true, false, true, false, true, true, true, true, true, true, true, false, false ]
[ null, "D", "D", "D", null, "D", "C", "D", "D", "D", "D", "D", "D", "D", null, null ]
[ { "extraction_type": "final", "non_overlapping_spans": null, "original_span": null, "span_end": null, "span_start": null, "total_spans_found": null }, { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "D", "span_end": 1326, "span_start": 1323,...
[ { "evaluation_method": "response_processor", "extraction_result": "{'extraction_type': 'final'}", "legacy_result": null, "reason": "no_extracted_answer" }, { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": true, "reason": null }, { "evaluation_m...
[ [], [ true ], [ true ], [ true ], [], [ true ], [ false ], [ true ], [ true ], [ true ], [ true ], [ true ], [ true ], [ true ], [], [] ]
[ [], [ "D" ], [ "D" ], [ "D" ], [], [ "D" ], [ "C" ], [ "D" ], [ "D" ], [ "D" ], [ "D" ], [ "D" ], [ "D" ], [ "D" ], [], [] ]
[ [], [ { "confidence": 1, "extraction_type": "internal", "original_span": "D", "pattern_used": "answer tag", "position": 1323, "span_end": 1326, "span_start": 1323, "span_text": "D" } ], [ { "confidence": 1, "extraction_type": "internal", ...
[ [], [ { "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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 11, "pass_at_n": 1, "percent_correct": 0.6875, "skill_count": { "answer_revision": [ 0, 0, 0, 0, 0, 0, ...
What is a wonderful place to get a cat?
humane society
commonsenseQA
csqa
[ { "content": "You like to solve problems by intentionally making mistakes then fixing them, then giving an answer. Think step by step, introduce an error, recognize it, fix it, finish out the rest of the reasoning, then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, y...
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "comfortable chair", "humane society", "walmart", "floor", "window sill" ] }
B
1
2630870837e16a36269dbb6ff87d42c2
[ "When thinking about where to get a cat, I want to consider a place that might not be immediately obvious for owning a pet. A comfortable chair seems too restrictive for a cat's natural needs. Walmart is also a common store where one might purchase food and supplies, but not necessarily a place to adopt from. The f...
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{ "api_url": "http://127.0.0.1:41177", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 4096, "n": 16, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vllm/Qw...
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ true, true, false, true, false, false, true, true, true, false, false, false, false, false, false, true ]
[ "B", "B", "HUMANE SOCIETY", "B", "D", "E", "B", "B", "B", "C", null, null, "D", "D", "H", "B" ]
[ { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "B", "span_end": 1121, "span_start": 1118, "total_spans_found": 1 }, { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "B", "span_end": 1270, "span_start": 1267, "t...
[ { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": true, "reason": null }, { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": true, "reason": null }, { "evaluation_method": "legacy", "extraction_result": null, "l...
[ [ true ], [ true ], [ false ], [ true ], [ false ], [ false ], [ true ], [ true ], [ true ], [ false ], [], [], [ false ], [ false ], [ false ], [ true ] ]
[ [ "B" ], [ "B" ], [ "HUMANE SOCIETY" ], [ "B" ], [ "D" ], [ "E" ], [ "B" ], [ "B" ], [ "B" ], [ "C" ], [], [], [ "D" ], [ "D" ], [ "H" ], [ "B" ] ]
[ [ { "confidence": 1, "extraction_type": "internal", "original_span": "B", "pattern_used": "answer tag", "position": 1118, "span_end": 1121, "span_start": 1118, "span_text": "B" } ], [ { "confidence": 1, "extraction_type": "internal", ...
[ [ { "evaluation_method": "legacy", "legacy_result": true } ], [ { "evaluation_method": "legacy", "legacy_result": true } ], [ { "evaluation_method": "legacy", "legacy_result": false } ], [ { "evaluation_method": "legacy", "legac...
{ "flips_by": [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 7, "pass_at_n": 1, "percent_correct": 0.4375, "skill_count": { "answer_revision": [ 0, 0, 0, 0, 0, 0, ...
The mom liked giving assistance to her children, why did she do it?
feel loved
commonsenseQA
csqa
[ { "content": "You like to solve problems by intentionally making mistakes then fixing them, then giving an answer. Think step by step, introduce an error, recognize it, fix it, finish out the rest of the reasoning, then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, y...
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "law suits", "feel loved", "contractual obligation", "happiness", "out of pocket" ] }
B
1
1e81653f9ab95af5a9e4f3e9385ad920
[ "Let's break down the scenario:\n\n1. **Identify the key elements**: The mom likes helping her children.\n2. **Consider possible reasons**:\n - A) Law suits (not typically related to parenting)\n - B) Feel loved (children often receive help from parents for emotional support)\n - C) Contractual obligation (th...
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{ "api_url": "http://127.0.0.1:41177", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 4096, "n": 16, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vllm/Qw...
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ true, false, true, false, true, true, false, false, true, false, false, false, true, false, false, true ]
[ "B", "D", "B", null, "B", "B", "D", "D", "B", "D", "D", "D", "B", null, "D", "B" ]
[ { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "B", "span_end": 1070, "span_start": 1069, "total_spans_found": 1 }, { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "D", "span_end": 1069, "span_start": 1066, "t...
[ { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": true, "reason": null }, { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": false, "reason": null }, { "evaluation_method": "legacy", "extraction_result": null, "...
[ [ true ], [ false ], [ true ], [], [ true ], [ true ], [ false ], [ false ], [ true ], [ false ], [ false ], [ false ], [ true ], [], [ false ], [ true ] ]
[ [ "B" ], [ "D" ], [ "B" ], [], [ "B" ], [ "B" ], [ "D" ], [ "D" ], [ "B" ], [ "D" ], [ "D" ], [ "D" ], [ "B" ], [], [ "D" ], [ "B" ] ]
[ [ { "confidence": 1, "extraction_type": "internal", "original_span": "B", "pattern_used": "answer tag", "position": 1069, "span_end": 1070, "span_start": 1069, "span_text": "B" } ], [ { "confidence": 1, "extraction_type": "internal", ...
[ [ { "evaluation_method": "legacy", "legacy_result": true } ], [ { "evaluation_method": "legacy", "legacy_result": false } ], [ { "evaluation_method": "legacy", "legacy_result": true } ], [], [ { "evaluation_method": "legacy", ...
{ "flips_by": [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 7, "pass_at_n": 1, "percent_correct": 0.4375, "skill_count": { "answer_revision": [ 0, 0, 0, 0, 0, 0, ...
where do some monkeys tend to live taht is hot?
tropical areas
commonsenseQA
csqa
[ { "content": "You like to solve problems by intentionally making mistakes then fixing them, then giving an answer. Think step by step, introduce an error, recognize it, fix it, finish out the rest of the reasoning, then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, y...
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "rain forest", "tree tops", "treetops", "great outdoors", "tropical areas" ] }
E
4
17364e82f8d105dea9875caf27fbfb45
[ "The correct answer would be A - rainforest.\nLet's break down why:\n\n1. **Intentional Mistake**: Rainforests are known for their high temperatures due to dense vegetation and heavy rainfall. This makes them ideal habitats for many warm-blooded animals, including monkeys.\n\n2. **Recognition of Error**: Monkeys ne...
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{ "api_url": "http://127.0.0.1:41177", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 4096, "n": 16, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vllm/Qw...
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, false, false, true, false, true, true, false, false, false, false, false, false, false, true, false ]
[ "A", null, "A", "E", "A", "E", "E", "A", "C", "C", null, "C", "C", "(B)", "E", "B" ]
[ { "extraction_type": "final", "non_overlapping_spans": 1, "original_span": "A", "span_end": 834, "span_start": 833, "total_spans_found": 1 }, { "extraction_type": "final", "non_overlapping_spans": null, "original_span": null, "span_end": null, "span_start": null, ...
[ { "evaluation_method": "legacy", "extraction_result": null, "legacy_result": false, "reason": null }, { "evaluation_method": "response_processor", "extraction_result": "{'extraction_type': 'final'}", "legacy_result": null, "reason": "no_extracted_answer" }, { "evaluation_...
[ [ false ], [], [ false ], [ true ], [ false ], [ true ], [ true ], [ false ], [ false ], [ false ], [], [ false ], [ false ], [ false ], [ true ], [ false ] ]
[ [ "A" ], [], [ "A" ], [ "E" ], [ "A" ], [ "E" ], [ "E" ], [ "A" ], [ "C" ], [ "C" ], [], [ "C" ], [ "C" ], [ "(B)" ], [ "E" ], [ "B" ] ]
[ [ { "confidence": 1, "extraction_type": "internal", "original_span": "A", "pattern_used": "answer tag", "position": 833, "span_end": 834, "span_start": 833, "span_text": "A" } ], [], [ { "confidence": 1, "extraction_type": "internal", ...
[ [ { "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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 4, "pass_at_n": 1, "percent_correct": 0.25, "skill_count": { "answer_revision": [ 0, 0, 0, 0, 0, 0, ...
What can I find that is about 4.5% alcohol?
beer
commonsenseQA
csqa
[{"content":"You like to solve problems by intentionally making mistakes then fixing them, then givi(...TRUNCATED)
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "beer", "wine", "pub", "uncle joe", "fraternity house" ] }
A
0
8e4f405dea84590ec6a0ce6d283c1997
["Let's approach this step-by-step:\n\n1. The question asks for something containing alcohol that ha(...TRUNCATED)
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:41177","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ true, false, false, false, false, true, false, true, false, false, true, false, false, true, true, false ]
[ "A", "(B)", "Uncle Joe", "B", "B", "A", "D", "A", null, "B", "A", null, "B", "A", "A", "beer" ]
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"A","span_end":1668,"span_star(...TRUNCATED)
[{"evaluation_method":"legacy","extraction_result":null,"legacy_result":true,"reason":null},{"evalua(...TRUNCATED)
[[true],[false],[false],[false],[false],[true],[false],[true],[],[false],[true],[],[false],[true],[t(...TRUNCATED)
[["A"],["(B)"],["Uncle Joe"],["B"],["B"],["A"],["D"],["A"],[],["B"],["A"],[],["B"],["A"],["A"],["bee(...TRUNCATED)
[[{"confidence":1.0,"extraction_type":"internal","original_span":"A","pattern_used":"answer tag","po(...TRUNCATED)
[[{"evaluation_method":"legacy","legacy_result":true}],[{"evaluation_method":"legacy","legacy_result(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":6,"pass_at_n":1,"percent(...TRUNCATED)
Where can we see an elephant do tricks?
circus
commonsenseQA
csqa
[{"content":"You like to solve problems by intentionally making mistakes then fixing them, then givi(...TRUNCATED)
[]
[]
backtrack
{ "label": [ "A", "B", "C", "D", "E" ], "text": [ "india", "circus", "zoo", "safari", "africa" ] }
B
1
366a0cde2e24373fae7b17ab37a6d1f6
["When solving this problem, let's first consider each option:\n\n1. **India**: While India has many(...TRUNCATED)
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:41177","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, true, true, true, false, true, true, true, true, true, true, true, false, false, true, true ]
[ "C", "B", "B", "B", "E", "B", "B", "B", "B", "B", "B", "B", "(No Valid Answer Exists)", null, "B", "B" ]
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"C","span_end":1030,"span_star(...TRUNCATED)
[{"evaluation_method":"legacy","extraction_result":null,"legacy_result":false,"reason":null},{"evalu(...TRUNCATED)
[[false],[true],[true],[true],[false],[true],[true],[true],[true],[true],[true],[true],[false],[],[t(...TRUNCATED)
[["C"],["B"],["B"],["B"],["E"],["B"],["B"],["B"],["B"],["B"],["B"],["B"],["(No Valid Answer Exists)"(...TRUNCATED)
[[{"confidence":1.0,"extraction_type":"internal","original_span":"C","pattern_used":"answer tag","po(...TRUNCATED)
[[{"evaluation_method":"legacy","legacy_result":false}],[{"evaluation_method":"legacy","legacy_resul(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":12,"pass_at_n":1,"percen(...TRUNCATED)
What can be used to remove a drop of blood?
needle
commonsenseQA
csqa
[{"content":"You like to solve problems by intentionally making mistakes then fixing them, then givi(...TRUNCATED)
[]
[]
backtrack
{"label":["A","B","C","D","E"],"text":["blood bank","needle","razor blade","slaughter house","person(...TRUNCATED)
B
1
509a8edb3497eeaa75faffc3158c8f1f
["Let's approach this step-by-step:\n\n1. The question asks what can be used to remove a drop of blo(...TRUNCATED)
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:41177","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, false, true, false, true, false, true, false, false, false, false, true, false, false, false, false ]
[ "E", "C", "B", "C", "B", "none of the above", "B", "person", null, "(E )", "E", "B", null, "None", "E", "person" ]
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"E","span_end":1725,"span_star(...TRUNCATED)
[{"evaluation_method":"legacy","extraction_result":null,"legacy_result":false,"reason":null},{"evalu(...TRUNCATED)
[[false],[false],[true],[false],[true],[false],[true],[false],[],[false],[false],[true],[],[false],[(...TRUNCATED)
[["E"],["C"],["B"],["C"],["B"],["none of the above"],["B"],["person"],[],["(E )"],["E"],["B"],[],["N(...TRUNCATED)
[[{"confidence":1.0,"extraction_type":"internal","original_span":"E","pattern_used":"answer tag","po(...TRUNCATED)
[[{"evaluation_method":"legacy","legacy_result":false}],[{"evaluation_method":"legacy","legacy_resul(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":4,"pass_at_n":1,"percent(...TRUNCATED)
What would someone do have fun at a winter party outside?
wrap up
commonsenseQA
csqa
[{"content":"You like to solve problems by intentionally making mistakes then fixing them, then givi(...TRUNCATED)
[]
[]
backtrack
{"label":["A","B","C","D","E"],"text":["find place to party","nothing in paticular","wrap up","drink(...TRUNCATED)
C
2
c2add4ee63f87acd2f22de06fecffee3
["When you're thinking about what someone might do at a winter party outside, the first thing that c(...TRUNCATED)
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:41177","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, false, false, false, false, false, false, false, false, true, false, false, false, false, false, true ]
[ "D", "A", "D", null, "B", "D", "A", "E", "D", "C", "E", "A", "E", "E", "D", "C" ]
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"D","span_end":1842,"span_star(...TRUNCATED)
[{"evaluation_method":"legacy","extraction_result":null,"legacy_result":false,"reason":null},{"evalu(...TRUNCATED)
[[false],[false],[false],[],[false],[false],[false],[false],[false],[true],[false],[false],[false],[(...TRUNCATED)
[ [ "D" ], [ "A" ], [ "D" ], [], [ "B" ], [ "D" ], [ "A" ], [ "E" ], [ "D" ], [ "C" ], [ "E" ], [ "A" ], [ "E" ], [ "E" ], [ "D" ], [ "C" ] ]
[[{"confidence":1.0,"extraction_type":"internal","original_span":"D","pattern_used":"answer tag","po(...TRUNCATED)
[[{"evaluation_method":"legacy","legacy_result":false}],[{"evaluation_method":"legacy","legacy_resul(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":2,"pass_at_n":1,"percent(...TRUNCATED)
"The catwalk was five stories up and exposed, everybody had to wear helmets and harnesses at the whe(...TRUNCATED)
construction site
commonsenseQA
csqa
[{"content":"You like to solve problems by intentionally making mistakes then fixing them, then givi(...TRUNCATED)
[]
[]
backtrack
{"label":["A","B","C","D","E"],"text":["mountain","construction site","theatre","fieldhouse","fashio(...TRUNCATED)
B
1
cdb10a509a23c76392482e1771a11d4e
["Let's approach this step-by-step:\n\n1. The catwalk is described as being \"five stories up and ex(...TRUNCATED)
[ false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:41177","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/Qwen/Qwen2.5-1.5B-Instruct", "n_responses": 16 }
[ false, false, true, false, true, false, false, true, false, false, true, true, false, false, false, false ]
[ "C", "C", "B", "C", "B", "D", null, "B", null, "E", "B", "B", "E", "A", "C", "C" ]
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"C","span_end":1884,"span_star(...TRUNCATED)
[{"evaluation_method":"legacy","extraction_result":null,"legacy_result":false,"reason":null},{"evalu(...TRUNCATED)
[[false],[false],[true],[false],[true],[false],[],[true],[],[false],[true],[true],[false],[false],[f(...TRUNCATED)
[ [ "C" ], [ "C" ], [ "B" ], [ "C" ], [ "B" ], [ "D" ], [], [ "B" ], [], [ "E" ], [ "B" ], [ "B" ], [ "E" ], [ "A" ], [ "C" ], [ "C" ] ]
[[{"confidence":1.0,"extraction_type":"internal","original_span":"C","pattern_used":"answer tag","po(...TRUNCATED)
[[{"evaluation_method":"legacy","legacy_result":false}],[{"evaluation_method":"legacy","legacy_resul(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":5,"pass_at_n":1,"percent(...TRUNCATED)
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
6