Datasets:

Modalities:
Text
Formats:
parquet
Size:
< 1K
Libraries:
Datasets
pandas
Dataset Viewer
Auto-converted to Parquet Duplicate
index
int64
0
29
words
listlengths
7
7
acronym
stringlengths
7
7
prompt
listlengths
2
2
task_source
stringclasses
1 value
model_responses__example_annotation
listlengths
8
8
model_responses__example_annotation__finish_reason_length_flags
listlengths
8
8
model_responses__example_annotation__length_partial_responses
listlengths
8
8
prompt__example_annotation__metadata
dict
model_responses__example_annotation__metadata
dict
model_responses__example_annotation__eval_is_correct
listlengths
8
8
model_responses__example_annotation__eval_extracted_answers
listlengths
8
8
model_responses__example_annotation__eval_extraction_metadata
listlengths
8
8
model_responses__example_annotation__eval_evaluation_metadata
listlengths
8
8
model_responses__example_annotation__internal_answers__eval_is_correct
listlengths
8
8
model_responses__example_annotation__internal_answers__eval_extracted_answers
listlengths
8
8
model_responses__example_annotation__internal_answers__eval_extraction_metadata
listlengths
8
8
model_responses__example_annotation__internal_answers__eval_evaluation_metadata
listlengths
8
8
model_responses__example_annotation__metrics
dict
0
[ "tpke", "giddying", "cassock", "overcall", "altamira", "confounds", "kinsfolk" ]
kgsvlck
[ { "content": "You like to solve problems by thinking step by step. Write out your reasoning then give your answer in \"<answer>\n(your answer)\n</answer>\". You always end with </answer>, you never ever end without giving an answer.", "role": "system" }, { "content": "Answer the following problem. E...
acronym
[ "<think>\n<sample>\nTo construct an acronym based on the given criteria, we need to follow these steps:\n\n1. **Identify the first three letters**: We start by taking the first three letters from any word in the list.\n2. **Check the remaining words**: Ensure that no other words contain those same initial letters a...
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{ "api_url": "http://127.0.0.1:56781", "backend": "litellm", "chat_template_applied": true, "generation_params": { "chat_template_applied": true, "max_tokens": 4096, "n": 3, "repetition_penalty": 1.1, "temperature": 0.7, "top_k": 20, "top_p": 0.8 }, "model_name": "hosted_vllm/TAU...
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[ "<answer>\n<<answer>\n</</the\">\"/>", null, "fokin", null, null, null, null, "tgeiocnfks" ]
[ { "extraction_type": "final", "non_overlapping_spans": 6, "original_span": "<answer>\n<<answer>\n</</the\">\"/>", "span_end": 13401, "span_start": 13369, "total_spans_found": 6 }, { "extraction_type": "final", "non_overlapping_spans": null, "original_span": null, "span_en...
[ { "correct_answer": null, "evaluation_method": "acronym_simple", "extracted_answer": "<answer>\n<<answer>\n</</the\">\"/>", "extraction_result": null, "reason": "no_metadata_available" }, { "correct_answer": null, "evaluation_method": "response_processor", "extracted_answer": nul...
[ [ false, false, false, false, false, false ], [], [ false ], [], [], [], [], [ false ] ]
[ [ "(incorrect)\n</\n\n \n\nLet us verify by kinship\nLet us finish here? \nkinfol\nhere\n\nHere ames\nkinfol\nkinold\nfol\n<answer>\n kin\nfollow\nkinfol\n<kin\nkinfol\n<villa>\n</kinfol\nkin\n<kin\nkin\nfollow\n<kin\n<kin\nkinfol\nkinfol\n<vael\nvaelher vael\n</\nkin\na\n<ale\n<a\n</answer\n (<answer>\n kin...
[ [ { "confidence": 1, "extraction_type": "internal", "original_span": "(incorrect)\n</\n\n \n\nLet us verify by kinship\nLet us finish here? \nkinfol\nhere\n\nHere ames\nkinfol\nkinold\nfol\n<answer>\n kin\nfollow\nkinfol\n<kin\nkinfol\n<villa>\n</kinfol\nkin\n<kin\nkin\nfollow\n<kin\n<kin\nk...
[ [ { "correct_answer": null, "evaluation_method": "acronym_simple", "extracted_answer": "(incorrect)\n</\n\n \n\nLet us verify by kinship\nLet us finish here? \nkinfol\nhere\n\nHere ames\nkinfol\nkinold\nfol\n<answer>\n kin\nfollow\nkinfol\n<kin\nkinfol\n<villa>\n</kinfol\nkin\n<kin\nkin\nfol...
{ "flips_by": [ 0, 0, 0, 0, 0, 0, 0, 0 ], "flips_total": 0, "num_correct": 0, "pass_at_n": 0, "percent_correct": 0, "skill_count": { "answer_revision": [ 0, 0, 0, 0, 0, 0, 0, 0 ], "best_of_n": [ 0, 0, ...
1
[ "emotionality", "directeur", "stephanion", "hurtful", "pendulum", "considerably", "nother" ]
ederent
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>\n<sample>\nTo construct an acronym based on the given requirements, we need to follow thes(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
["</ in instead. that then a's own that to that the the the answer that answer is :then that a that (...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":3,"original_span":"</ in instead. that then a's (...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"</ in instead. that(...TRUNCATED)
[ [ false, false, false ], [ false, false, false, false, false ], [], [], [], [], [ false ], [ false ] ]
[["dspi","entryman entryman </ entryman </answer is theronym </answer]","</ in instead. that then a'(...TRUNCATED)
[[{"confidence":1.0,"extraction_type":"internal","original_span":"dspi","pattern_used":"answer tag",(...TRUNCATED)
[[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"dspi","reason":"no(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
2
[ "underscript", "extrasterile", "beslime", "emotionist", "soloth", "alsike", "event" ]
nteelae
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>\n<sample>\nTo construct an acronym based on the given requirements, we need to follow thes(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[". have. . </answer. . have. </answer. </answer. >. have. </ answer. </ answer. </ have. </ have. <(...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":". have. . </answer. . have. <(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":". have. . </answer.(...TRUNCATED)
[ [ false ], [ false ], [ false ], [ false ], [], [], [ false ], [] ]
[[". have. . </answer. . have. </answer. </answer. >. have. </ answer. </ answer. </ have. </ have. (...TRUNCATED)
[[{"confidence":1.0,"extraction_type":"internal","original_span":". have. . </answer. . have. </answ(...TRUNCATED)
[[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":". have. . </answer(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
3
[ "pasmo", "residual", "trochius", "scats", "yird", "cateye", "noneugenical" ]
psrcico
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>\n<sample>\nTo construct an acronym based on the given requirements, we need to follow thes(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[null,"</\n Answer the<spanAnswer< answer</ Answer Answer the``< Answer>.</ Answer Answer< Answer/",(...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"response_processor","extracted_answer":null,"extraction(...TRUNCATED)
[ [], [ false, false, false ], [ false, false ], [], [], [ false, false, false ], [], [ false, false, false, false ] ]
[[],[">: , the\n the you was not found the.\n answer\n\n is is<answer\n\n<answer > taken take t the (...TRUNCATED)
[[],[{"confidence":1.0,"extraction_type":"internal","original_span":">: , the\n the you was not foun(...TRUNCATED)
[[],[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":">: , the\n the (...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
4
[ "cutters", "caseum", "pachycholia", "vorticellum", "albiflorous", "reflectorize", "hypothecater" ]
ccpvaeh
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>\n<sample>\nTo construct an acronym based on the given requirements, let's break down the p(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[null,null,". < Be. > have. Be. A. have. A. Have. Be. A. have.. <answer.\n:\n. andelson. </answer. <(...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"response_processor","extracted_answer":null,"extraction(...TRUNCATED)
[ [], [], [ false ], [ false, false ], [ false ], [], [ false ], [] ]
[[],[],[". < Be. > have. Be. A. have. A. Have. Be. A. have.. <answer.\n:\n. andelson. </answer. </an(...TRUNCATED)
[[],[],[{"confidence":1.0,"extraction_type":"internal","original_span":". < Be. > have. Be. A. have.(...TRUNCATED)
[[],[],[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":". < Be. > ha(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
5
[ "impanated", "sprayful", "nondeformed", "phorone", "fleckering", "plutarchic", "casate" ]
psohfps
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>>\n\n> nonpe, etcs for that word's. \" the\n\nthe next the word\n\n \". word] a\". now !/ (...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[ "<answer> ( (your answer)", "</\nif .<answer>\n</ \nanswer.\nthe", null, null, null, null, null, null ]
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"<answer> ( (your answer)","sp(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"<answer> ( (your an(...TRUNCATED)
[ [ false ], [ false, false ], [], [], [], [], [], [] ]
[ [ "<answer> ( (your answer)" ], [ "</", "</\nif .<answer>\n</ \nanswer.\nthe" ], [], [], [], [], [], [] ]
[[{"confidence":1.0,"extraction_type":"internal","original_span":"<answer> ( (your answer)","pattern(...TRUNCATED)
[[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"<answer> ( (your a(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
6
[ "zymogen", "mitigation", "kamikazes", "acron", "amydon", "semiputrid", "unfrozen" ]
yimamen
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>>\n\n we> consider. \n \n\nStep the answer the]( \nthe. \n .\n the to construct. the(cc\(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[ null, "( your </ answer </ answer>", null, null, null, null, null, null ]
[{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"response_processor","extracted_answer":null,"extraction(...TRUNCATED)
[ [], [ false, false, false ], [], [], [], [], [], [] ]
[ [], [ "(your answer)", "( (your answer) </answer", "( your </ answer </ answer>" ], [], [], [], [], [], [] ]
[[],[{"confidence":1.0,"extraction_type":"internal","original_span":"(your answer)","pattern_used":"(...TRUNCATED)
[[],[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"(your answer)",(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
7
[ "skiagrams", "manifoldly", "sidecars", "azotes", "contendere", "raid", "dewworm" ]
knizoid
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>\n\n</\n\n( your youthen\n (your answer) no uian. Let th . </ >)</answer>>then then a (...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[null,"</ >","the\n\n</answer\n>\n </answer\n\na</</</answer\n<span> on the<span class=\"math-t clas(...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"response_processor","extracted_answer":null,"extraction(...TRUNCATED)
[[],[false,false,false,false,false,false,false,false],[false,false],[false],[false],[false,false],[f(...TRUNCATED)
[[],["> . will to see just, and you< sideav<answer>\n>\n (\n I have been<document> to write.</ the w(...TRUNCATED)
[[],[{"confidence":1.0,"extraction_type":"internal","original_span":"> . will to see just, and you< (...TRUNCATED)
[[],[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"> . will to see(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
8
[ "jacksmelt", "aediles", "distender", "preovulatory", "gussies", "latticinii", "bootblacks" ]
aadputo
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<thadow</sadow\nanswerOP into> before\n\nand\n ov pre before though. first > \" >\n </a else if.s(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
[null,"(rather than using letters).</see so. < .. )(but)",null,"</the?\n we Now let's break our\n no(...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":null,"original_span":null,"span_end":null,"span_(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"response_processor","extracted_answer":null,"extraction(...TRUNCATED)
[ [], [ false, false ], [], [ false ], [ false ], [ false, false, false, false, false, false ], [ false, false, false ], [] ]
[[],["j the. </. </. <answer>the letter \" has…</iseand id. <answer>\na. \n \n<The. <Ifas been t(...TRUNCATED)
[[],[{"confidence":1.0,"extraction_type":"internal","original_span":"j the. </. </. <answer>the lett(...TRUNCATED)
[[],[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"j the. </. </. (...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
9
[ "brocade", "lightwort", "upstruggle", "tortrixes", "philadelphus", "tim", "uptrill" ]
rgsrpiu
[{"content":"You like to solve problems by thinking step by step. Write out your reasoning then give(...TRUNCATED)
acronym
["<think>\n<answer>\n\nthinker\nthe...than off in ...see to write now \n appear to x\n<answer>\nasse(...TRUNCATED)
[ false, false, false, false, false, false, false, false ]
[ "", "", "", "", "", "", "", "" ]
{"api_url":"http://127.0.0.1:56781","backend":"litellm","chat_template_applied":true,"generation_par(...TRUNCATED)
{ "backend": "litellm", "model_name": "hosted_vllm/TAUR-dev/M-SFT_V2_RUN-rl", "n_responses": 8 }
[ false, false, false, false, false, false, false, false ]
["thinker\nthe...than off in ...see to write now \n appear to x\n<answer>\nasser. begin\n\n. . . .\n(...TRUNCATED)
[{"extraction_type":"final","non_overlapping_spans":1,"original_span":"thinker\nthe...than off in ..(...TRUNCATED)
[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"thinker\nthe...than(...TRUNCATED)
[ [ false ], [], [], [], [ false, false ], [ false ], [], [] ]
[["thinker\nthe...than off in ...see to write now \n appear to x\n<answer>\nasser. begin\n\n. . . .\(...TRUNCATED)
[[{"confidence":1.0,"extraction_type":"internal","original_span":"thinker\nthe...than off in ...see (...TRUNCATED)
[[{"correct_answer":null,"evaluation_method":"acronym_simple","extracted_answer":"thinker\nthe...tha(...TRUNCATED)
{"flips_by":[0,0,0,0,0,0,0,0],"flips_total":0,"num_correct":0,"pass_at_n":0,"percent_correct":0.0,"s(...TRUNCATED)
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
6