| { |
| "results": { |
| "molestiae-aperiam_lsat-rc_cot": { |
| "acc,none": 0.4758364312267658, |
| "acc_stderr,none": 0.030506674211283072, |
| "alias": "molestiae-aperiam_lsat-rc_cot" |
| }, |
| "molestiae-aperiam_lsat-lr_cot": { |
| "acc,none": 0.4196078431372549, |
| "acc_stderr,none": 0.021873771696750574, |
| "alias": "molestiae-aperiam_lsat-lr_cot" |
| }, |
| "molestiae-aperiam_lsat-ar_cot": { |
| "acc,none": 0.24347826086956523, |
| "acc_stderr,none": 0.02836109930007507, |
| "alias": "molestiae-aperiam_lsat-ar_cot" |
| }, |
| "molestiae-aperiam_logiqa_cot": { |
| "acc,none": 0.3753993610223642, |
| "acc_stderr,none": 0.019369034287392513, |
| "alias": "molestiae-aperiam_logiqa_cot" |
| }, |
| "molestiae-aperiam_logiqa2_cot": { |
| "acc,none": 0.477735368956743, |
| "acc_stderr,none": 0.012602331890057498, |
| "alias": "molestiae-aperiam_logiqa2_cot" |
| }, |
| "iure-at_lsat-rc_cot": { |
| "acc,none": 0.5241635687732342, |
| "acc_stderr,none": 0.03050667421128308, |
| "alias": "iure-at_lsat-rc_cot" |
| }, |
| "iure-at_lsat-lr_cot": { |
| "acc,none": 0.4666666666666667, |
| "acc_stderr,none": 0.02211280638156422, |
| "alias": "iure-at_lsat-lr_cot" |
| }, |
| "iure-at_lsat-ar_cot": { |
| "acc,none": 0.19130434782608696, |
| "acc_stderr,none": 0.025991852462828483, |
| "alias": "iure-at_lsat-ar_cot" |
| }, |
| "iure-at_logiqa_cot": { |
| "acc,none": 0.36421725239616615, |
| "acc_stderr,none": 0.019248399225001777, |
| "alias": "iure-at_logiqa_cot" |
| }, |
| "iure-at_logiqa2_cot": { |
| "acc,none": 0.4446564885496183, |
| "acc_stderr,none": 0.012537330526916487, |
| "alias": "iure-at_logiqa2_cot" |
| }, |
| "facere-optio_lsat-rc_cot": { |
| "acc,none": 0.5018587360594795, |
| "acc_stderr,none": 0.030542150046756426, |
| "alias": "facere-optio_lsat-rc_cot" |
| }, |
| "facere-optio_lsat-lr_cot": { |
| "acc,none": 0.43333333333333335, |
| "acc_stderr,none": 0.021964230412067975, |
| "alias": "facere-optio_lsat-lr_cot" |
| }, |
| "facere-optio_lsat-ar_cot": { |
| "acc,none": 0.1826086956521739, |
| "acc_stderr,none": 0.02553042195273417, |
| "alias": "facere-optio_lsat-ar_cot" |
| }, |
| "facere-optio_logiqa_cot": { |
| "acc,none": 0.3706070287539936, |
| "acc_stderr,none": 0.019318693909977667, |
| "alias": "facere-optio_logiqa_cot" |
| }, |
| "facere-optio_logiqa2_cot": { |
| "acc,none": 0.44656488549618323, |
| "acc_stderr,none": 0.012542599303456, |
| "alias": "facere-optio_logiqa2_cot" |
| }, |
| "et-praesentium_lsat-rc_cot": { |
| "acc,none": 0.49814126394052044, |
| "acc_stderr,none": 0.030542150046756422, |
| "alias": "et-praesentium_lsat-rc_cot" |
| }, |
| "et-praesentium_lsat-lr_cot": { |
| "acc,none": 0.3784313725490196, |
| "acc_stderr,none": 0.02149706741180824, |
| "alias": "et-praesentium_lsat-lr_cot" |
| }, |
| "et-praesentium_lsat-ar_cot": { |
| "acc,none": 0.2217391304347826, |
| "acc_stderr,none": 0.02745149660405892, |
| "alias": "et-praesentium_lsat-ar_cot" |
| }, |
| "et-praesentium_logiqa_cot": { |
| "acc,none": 0.36261980830670926, |
| "acc_stderr,none": 0.019230254618400246, |
| "alias": "et-praesentium_logiqa_cot" |
| }, |
| "et-praesentium_logiqa2_cot": { |
| "acc,none": 0.46946564885496184, |
| "acc_stderr,none": 0.012591300013425958, |
| "alias": "et-praesentium_logiqa2_cot" |
| }, |
| "eligendi-commodi_lsat-rc_cot": { |
| "acc,none": 0.5018587360594795, |
| "acc_stderr,none": 0.030542150046756426, |
| "alias": "eligendi-commodi_lsat-rc_cot" |
| }, |
| "eligendi-commodi_lsat-lr_cot": { |
| "acc,none": 0.42549019607843136, |
| "acc_stderr,none": 0.021914653579107275, |
| "alias": "eligendi-commodi_lsat-lr_cot" |
| }, |
| "eligendi-commodi_lsat-ar_cot": { |
| "acc,none": 0.23478260869565218, |
| "acc_stderr,none": 0.02800964707093011, |
| "alias": "eligendi-commodi_lsat-ar_cot" |
| }, |
| "eligendi-commodi_logiqa_cot": { |
| "acc,none": 0.3466453674121406, |
| "acc_stderr,none": 0.019036064999420094, |
| "alias": "eligendi-commodi_logiqa_cot" |
| }, |
| "eligendi-commodi_logiqa2_cot": { |
| "acc,none": 0.4643765903307888, |
| "acc_stderr,none": 0.012582786901750204, |
| "alias": "eligendi-commodi_logiqa2_cot" |
| }, |
| "doloremque-rem_lsat-rc_cot": { |
| "acc,none": 0.5092936802973977, |
| "acc_stderr,none": 0.030537084593525398, |
| "alias": "doloremque-rem_lsat-rc_cot" |
| }, |
| "doloremque-rem_lsat-lr_cot": { |
| "acc,none": 0.4117647058823529, |
| "acc_stderr,none": 0.02181429628344194, |
| "alias": "doloremque-rem_lsat-lr_cot" |
| }, |
| "doloremque-rem_lsat-ar_cot": { |
| "acc,none": 0.21304347826086956, |
| "acc_stderr,none": 0.027057754389936194, |
| "alias": "doloremque-rem_lsat-ar_cot" |
| }, |
| "doloremque-rem_logiqa_cot": { |
| "acc,none": 0.3610223642172524, |
| "acc_stderr,none": 0.019211880355748154, |
| "alias": "doloremque-rem_logiqa_cot" |
| }, |
| "doloremque-rem_logiqa2_cot": { |
| "acc,none": 0.46119592875318066, |
| "acc_stderr,none": 0.012576797669813296, |
| "alias": "doloremque-rem_logiqa2_cot" |
| } |
| }, |
| "configs": { |
| "doloremque-rem_logiqa2_cot": { |
| "task": "doloremque-rem_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "doloremque-rem-logiqa2/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "doloremque-rem_logiqa_cot": { |
| "task": "doloremque-rem_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "doloremque-rem-logiqa/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "doloremque-rem_lsat-ar_cot": { |
| "task": "doloremque-rem_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "doloremque-rem-lsat-ar/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "doloremque-rem_lsat-lr_cot": { |
| "task": "doloremque-rem_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "doloremque-rem-lsat-lr/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "doloremque-rem_lsat-rc_cot": { |
| "task": "doloremque-rem_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "doloremque-rem-lsat-rc/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "eligendi-commodi_logiqa2_cot": { |
| "task": "eligendi-commodi_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eligendi-commodi-logiqa2/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "eligendi-commodi_logiqa_cot": { |
| "task": "eligendi-commodi_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eligendi-commodi-logiqa/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "eligendi-commodi_lsat-ar_cot": { |
| "task": "eligendi-commodi_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eligendi-commodi-lsat-ar/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "eligendi-commodi_lsat-lr_cot": { |
| "task": "eligendi-commodi_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eligendi-commodi-lsat-lr/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "eligendi-commodi_lsat-rc_cot": { |
| "task": "eligendi-commodi_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eligendi-commodi-lsat-rc/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "et-praesentium_logiqa2_cot": { |
| "task": "et-praesentium_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "et-praesentium-logiqa2/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "et-praesentium_logiqa_cot": { |
| "task": "et-praesentium_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "et-praesentium-logiqa/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "et-praesentium_lsat-ar_cot": { |
| "task": "et-praesentium_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "et-praesentium-lsat-ar/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "et-praesentium_lsat-lr_cot": { |
| "task": "et-praesentium_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "et-praesentium-lsat-lr/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "et-praesentium_lsat-rc_cot": { |
| "task": "et-praesentium_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "et-praesentium-lsat-rc/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "facere-optio_logiqa2_cot": { |
| "task": "facere-optio_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "facere-optio-logiqa2/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "facere-optio_logiqa_cot": { |
| "task": "facere-optio_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "facere-optio-logiqa/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "facere-optio_lsat-ar_cot": { |
| "task": "facere-optio_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "facere-optio-lsat-ar/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "facere-optio_lsat-lr_cot": { |
| "task": "facere-optio_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "facere-optio-lsat-lr/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "facere-optio_lsat-rc_cot": { |
| "task": "facere-optio_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "facere-optio-lsat-rc/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "iure-at_logiqa2_cot": { |
| "task": "iure-at_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iure-at-logiqa2/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "iure-at_logiqa_cot": { |
| "task": "iure-at_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iure-at-logiqa/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "iure-at_lsat-ar_cot": { |
| "task": "iure-at_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iure-at-lsat-ar/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "iure-at_lsat-lr_cot": { |
| "task": "iure-at_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iure-at-lsat-lr/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "iure-at_lsat-rc_cot": { |
| "task": "iure-at_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iure-at-lsat-rc/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "molestiae-aperiam_logiqa2_cot": { |
| "task": "molestiae-aperiam_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "molestiae-aperiam-logiqa2/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "molestiae-aperiam_logiqa_cot": { |
| "task": "molestiae-aperiam_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "molestiae-aperiam-logiqa/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "molestiae-aperiam_lsat-ar_cot": { |
| "task": "molestiae-aperiam_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "molestiae-aperiam-lsat-ar/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "molestiae-aperiam_lsat-lr_cot": { |
| "task": "molestiae-aperiam_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "molestiae-aperiam-lsat-lr/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| }, |
| "molestiae-aperiam_lsat-rc_cot": { |
| "task": "molestiae-aperiam_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "cot-leaderboard/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "molestiae-aperiam-lsat-rc/test-00000-of-00001.parquet" |
| } |
| }, |
| "test_split": "test", |
| "doc_to_text": "def doc_to_text_cot(doc) -> str:\n \"\"\"\n Answer the following question about the given passage. [Base your answer on the reasoning below.]\n \n Passage: <passage>\n \n Question: <question>\n A. <choice1>\n B. <choice2>\n C. <choice3>\n D. <choice4>\n [E. <choice5>]\n \n [Reasoning: <reasoning>]\n \n Answer:\n \"\"\"\n k = len(doc[\"options\"])\n choices = [\"a\", \"b\", \"c\", \"d\", \"e\"][:k]\n prompt = \"Answer the following question about the given passage. Base your answer on the reasoning below.\\n\\n\"\n prompt = \"Passage: \" + doc[\"passage\"] + \"\\n\\n\"\n prompt += \"Question: \" + doc[\"question\"] + \"\\n\"\n for choice, option in zip(choices, doc[\"options\"]):\n prompt += f\"{choice.upper()}. {option}\\n\"\n prompt += \"\\n\"\n prompt += \"Reasoning: \" + doc[\"reasoning_trace\"] + \"\\n\\n\" \n prompt += \"Answer:\"\n return prompt\n", |
| "doc_to_target": "{{answer}}", |
| "doc_to_choice": "{{options}}", |
| "description": "", |
| "target_delimiter": " ", |
| "fewshot_delimiter": "\n\n", |
| "num_fewshot": 0, |
| "metric_list": [ |
| { |
| "metric": "acc", |
| "aggregation": "mean", |
| "higher_is_better": true |
| } |
| ], |
| "output_type": "multiple_choice", |
| "repeats": 1, |
| "should_decontaminate": false, |
| "metadata": { |
| "version": 0.0 |
| } |
| } |
| }, |
| "versions": { |
| "doloremque-rem_logiqa2_cot": 0.0, |
| "doloremque-rem_logiqa_cot": 0.0, |
| "doloremque-rem_lsat-ar_cot": 0.0, |
| "doloremque-rem_lsat-lr_cot": 0.0, |
| "doloremque-rem_lsat-rc_cot": 0.0, |
| "eligendi-commodi_logiqa2_cot": 0.0, |
| "eligendi-commodi_logiqa_cot": 0.0, |
| "eligendi-commodi_lsat-ar_cot": 0.0, |
| "eligendi-commodi_lsat-lr_cot": 0.0, |
| "eligendi-commodi_lsat-rc_cot": 0.0, |
| "et-praesentium_logiqa2_cot": 0.0, |
| "et-praesentium_logiqa_cot": 0.0, |
| "et-praesentium_lsat-ar_cot": 0.0, |
| "et-praesentium_lsat-lr_cot": 0.0, |
| "et-praesentium_lsat-rc_cot": 0.0, |
| "facere-optio_logiqa2_cot": 0.0, |
| "facere-optio_logiqa_cot": 0.0, |
| "facere-optio_lsat-ar_cot": 0.0, |
| "facere-optio_lsat-lr_cot": 0.0, |
| "facere-optio_lsat-rc_cot": 0.0, |
| "iure-at_logiqa2_cot": 0.0, |
| "iure-at_logiqa_cot": 0.0, |
| "iure-at_lsat-ar_cot": 0.0, |
| "iure-at_lsat-lr_cot": 0.0, |
| "iure-at_lsat-rc_cot": 0.0, |
| "molestiae-aperiam_logiqa2_cot": 0.0, |
| "molestiae-aperiam_logiqa_cot": 0.0, |
| "molestiae-aperiam_lsat-ar_cot": 0.0, |
| "molestiae-aperiam_lsat-lr_cot": 0.0, |
| "molestiae-aperiam_lsat-rc_cot": 0.0 |
| }, |
| "n-shot": { |
| "doloremque-rem_logiqa2_cot": 0, |
| "doloremque-rem_logiqa_cot": 0, |
| "doloremque-rem_lsat-ar_cot": 0, |
| "doloremque-rem_lsat-lr_cot": 0, |
| "doloremque-rem_lsat-rc_cot": 0, |
| "eligendi-commodi_logiqa2_cot": 0, |
| "eligendi-commodi_logiqa_cot": 0, |
| "eligendi-commodi_lsat-ar_cot": 0, |
| "eligendi-commodi_lsat-lr_cot": 0, |
| "eligendi-commodi_lsat-rc_cot": 0, |
| "et-praesentium_logiqa2_cot": 0, |
| "et-praesentium_logiqa_cot": 0, |
| "et-praesentium_lsat-ar_cot": 0, |
| "et-praesentium_lsat-lr_cot": 0, |
| "et-praesentium_lsat-rc_cot": 0, |
| "facere-optio_logiqa2_cot": 0, |
| "facere-optio_logiqa_cot": 0, |
| "facere-optio_lsat-ar_cot": 0, |
| "facere-optio_lsat-lr_cot": 0, |
| "facere-optio_lsat-rc_cot": 0, |
| "iure-at_logiqa2_cot": 0, |
| "iure-at_logiqa_cot": 0, |
| "iure-at_lsat-ar_cot": 0, |
| "iure-at_lsat-lr_cot": 0, |
| "iure-at_lsat-rc_cot": 0, |
| "molestiae-aperiam_logiqa2_cot": 0, |
| "molestiae-aperiam_logiqa_cot": 0, |
| "molestiae-aperiam_lsat-ar_cot": 0, |
| "molestiae-aperiam_lsat-lr_cot": 0, |
| "molestiae-aperiam_lsat-rc_cot": 0 |
| }, |
| "config": { |
| "model": "vllm", |
| "model_args": "pretrained=openchat/openchat-3.5-0106,revision=main,dtype=auto,tensor_parallel_size=1,gpu_memory_utilization=0.9,trust_remote_code=true,max_length=4096", |
| "batch_size": "auto", |
| "batch_sizes": [], |
| "device": null, |
| "use_cache": null, |
| "limit": null, |
| "bootstrap_iters": 100000, |
| "gen_kwargs": null |
| }, |
| "git_hash": "a1d6b70" |
| } |
