| { |
| "results": { |
| "veritatis-velit_lsat-rc_cot": { |
| "acc,none": 0.4163568773234201, |
| "acc_stderr,none": 0.03011196940753653, |
| "alias": "veritatis-velit_lsat-rc_cot" |
| }, |
| "veritatis-velit_lsat-lr_cot": { |
| "acc,none": 0.3137254901960784, |
| "acc_stderr,none": 0.02056671577177923, |
| "alias": "veritatis-velit_lsat-lr_cot" |
| }, |
| "veritatis-velit_lsat-ar_cot": { |
| "acc,none": 0.23043478260869565, |
| "acc_stderr,none": 0.027827807522276156, |
| "alias": "veritatis-velit_lsat-ar_cot" |
| }, |
| "veritatis-velit_logiqa_cot": { |
| "acc,none": 0.2987220447284345, |
| "acc_stderr,none": 0.01830790800596066, |
| "alias": "veritatis-velit_logiqa_cot" |
| }, |
| "veritatis-velit_logiqa2_cot": { |
| "acc,none": 0.3530534351145038, |
| "acc_stderr,none": 0.012057751628201937, |
| "alias": "veritatis-velit_logiqa2_cot" |
| }, |
| "saepe-fuga_lsat-rc_cot": { |
| "acc,none": 0.4312267657992565, |
| "acc_stderr,none": 0.030252065567665472, |
| "alias": "saepe-fuga_lsat-rc_cot" |
| }, |
| "saepe-fuga_lsat-lr_cot": { |
| "acc,none": 0.296078431372549, |
| "acc_stderr,none": 0.020235159438512108, |
| "alias": "saepe-fuga_lsat-lr_cot" |
| }, |
| "saepe-fuga_lsat-ar_cot": { |
| "acc,none": 0.18695652173913044, |
| "acc_stderr,none": 0.02576377239851234, |
| "alias": "saepe-fuga_lsat-ar_cot" |
| }, |
| "saepe-fuga_logiqa_cot": { |
| "acc,none": 0.28913738019169327, |
| "acc_stderr,none": 0.018134473494097365, |
| "alias": "saepe-fuga_logiqa_cot" |
| }, |
| "saepe-fuga_logiqa2_cot": { |
| "acc,none": 0.356234096692112, |
| "acc_stderr,none": 0.012082133651061318, |
| "alias": "saepe-fuga_logiqa2_cot" |
| }, |
| "nisi-sunt_lsat-rc_cot": { |
| "acc,none": 0.34572490706319703, |
| "acc_stderr,none": 0.029052140190085934, |
| "alias": "nisi-sunt_lsat-rc_cot" |
| }, |
| "nisi-sunt_lsat-lr_cot": { |
| "acc,none": 0.27647058823529413, |
| "acc_stderr,none": 0.019824108780753004, |
| "alias": "nisi-sunt_lsat-lr_cot" |
| }, |
| "nisi-sunt_lsat-ar_cot": { |
| "acc,none": 0.27391304347826084, |
| "acc_stderr,none": 0.029470189815005897, |
| "alias": "nisi-sunt_lsat-ar_cot" |
| }, |
| "nisi-sunt_logiqa_cot": { |
| "acc,none": 0.31629392971246006, |
| "acc_stderr,none": 0.018601164683514252, |
| "alias": "nisi-sunt_logiqa_cot" |
| }, |
| "nisi-sunt_logiqa2_cot": { |
| "acc,none": 0.3708651399491094, |
| "acc_stderr,none": 0.012186859070473788, |
| "alias": "nisi-sunt_logiqa2_cot" |
| }, |
| "laboriosam-molestiae_lsat-rc_cot": { |
| "acc,none": 0.3680297397769517, |
| "acc_stderr,none": 0.029459297142360178, |
| "alias": "laboriosam-molestiae_lsat-rc_cot" |
| }, |
| "laboriosam-molestiae_lsat-lr_cot": { |
| "acc,none": 0.2823529411764706, |
| "acc_stderr,none": 0.019952288758197854, |
| "alias": "laboriosam-molestiae_lsat-lr_cot" |
| }, |
| "laboriosam-molestiae_lsat-ar_cot": { |
| "acc,none": 0.2217391304347826, |
| "acc_stderr,none": 0.027451496604058916, |
| "alias": "laboriosam-molestiae_lsat-ar_cot" |
| }, |
| "laboriosam-molestiae_logiqa_cot": { |
| "acc,none": 0.3083067092651757, |
| "acc_stderr,none": 0.018471759300608265, |
| "alias": "laboriosam-molestiae_logiqa_cot" |
| }, |
| "laboriosam-molestiae_logiqa2_cot": { |
| "acc,none": 0.36895674300254455, |
| "acc_stderr,none": 0.012173885104839207, |
| "alias": "laboriosam-molestiae_logiqa2_cot" |
| }, |
| "iste-molestias_lsat-rc_cot": { |
| "acc,none": 0.4275092936802974, |
| "acc_stderr,none": 0.030219662071838058, |
| "alias": "iste-molestias_lsat-rc_cot" |
| }, |
| "iste-molestias_lsat-lr_cot": { |
| "acc,none": 0.2647058823529412, |
| "acc_stderr,none": 0.01955480325785009, |
| "alias": "iste-molestias_lsat-lr_cot" |
| }, |
| "iste-molestias_lsat-ar_cot": { |
| "acc,none": 0.20434782608695654, |
| "acc_stderr,none": 0.02664580815001135, |
| "alias": "iste-molestias_lsat-ar_cot" |
| }, |
| "iste-molestias_logiqa_cot": { |
| "acc,none": 0.3003194888178914, |
| "acc_stderr,none": 0.018335874932123606, |
| "alias": "iste-molestias_logiqa_cot" |
| }, |
| "iste-molestias_logiqa2_cot": { |
| "acc,none": 0.3505089058524173, |
| "acc_stderr,none": 0.012037825298569541, |
| "alias": "iste-molestias_logiqa2_cot" |
| }, |
| "eum-saepe_lsat-rc_cot": { |
| "acc,none": 0.44609665427509293, |
| "acc_stderr,none": 0.030364356394504122, |
| "alias": "eum-saepe_lsat-rc_cot" |
| }, |
| "eum-saepe_lsat-lr_cot": { |
| "acc,none": 0.2803921568627451, |
| "acc_stderr,none": 0.019910033171474082, |
| "alias": "eum-saepe_lsat-lr_cot" |
| }, |
| "eum-saepe_lsat-ar_cot": { |
| "acc,none": 0.23478260869565218, |
| "acc_stderr,none": 0.028009647070930125, |
| "alias": "eum-saepe_lsat-ar_cot" |
| }, |
| "eum-saepe_logiqa_cot": { |
| "acc,none": 0.30670926517571884, |
| "acc_stderr,none": 0.018445105229565346, |
| "alias": "eum-saepe_logiqa_cot" |
| }, |
| "eum-saepe_logiqa2_cot": { |
| "acc,none": 0.36323155216284986, |
| "acc_stderr,none": 0.012133733683836157, |
| "alias": "eum-saepe_logiqa2_cot" |
| } |
| }, |
| "configs": { |
| "eum-saepe_logiqa2_cot": { |
| "task": "eum-saepe_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eum-saepe-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 |
| } |
| }, |
| "eum-saepe_logiqa_cot": { |
| "task": "eum-saepe_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eum-saepe-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 |
| } |
| }, |
| "eum-saepe_lsat-ar_cot": { |
| "task": "eum-saepe_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eum-saepe-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 |
| } |
| }, |
| "eum-saepe_lsat-lr_cot": { |
| "task": "eum-saepe_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eum-saepe-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 |
| } |
| }, |
| "eum-saepe_lsat-rc_cot": { |
| "task": "eum-saepe_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "eum-saepe-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 |
| } |
| }, |
| "iste-molestias_logiqa2_cot": { |
| "task": "iste-molestias_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iste-molestias-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 |
| } |
| }, |
| "iste-molestias_logiqa_cot": { |
| "task": "iste-molestias_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iste-molestias-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 |
| } |
| }, |
| "iste-molestias_lsat-ar_cot": { |
| "task": "iste-molestias_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iste-molestias-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 |
| } |
| }, |
| "iste-molestias_lsat-lr_cot": { |
| "task": "iste-molestias_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iste-molestias-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 |
| } |
| }, |
| "iste-molestias_lsat-rc_cot": { |
| "task": "iste-molestias_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "iste-molestias-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 |
| } |
| }, |
| "laboriosam-molestiae_logiqa2_cot": { |
| "task": "laboriosam-molestiae_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "laboriosam-molestiae-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 |
| } |
| }, |
| "laboriosam-molestiae_logiqa_cot": { |
| "task": "laboriosam-molestiae_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "laboriosam-molestiae-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 |
| } |
| }, |
| "laboriosam-molestiae_lsat-ar_cot": { |
| "task": "laboriosam-molestiae_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "laboriosam-molestiae-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 |
| } |
| }, |
| "laboriosam-molestiae_lsat-lr_cot": { |
| "task": "laboriosam-molestiae_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "laboriosam-molestiae-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 |
| } |
| }, |
| "laboriosam-molestiae_lsat-rc_cot": { |
| "task": "laboriosam-molestiae_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "laboriosam-molestiae-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 |
| } |
| }, |
| "nisi-sunt_logiqa2_cot": { |
| "task": "nisi-sunt_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "nisi-sunt-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 |
| } |
| }, |
| "nisi-sunt_logiqa_cot": { |
| "task": "nisi-sunt_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "nisi-sunt-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 |
| } |
| }, |
| "nisi-sunt_lsat-ar_cot": { |
| "task": "nisi-sunt_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "nisi-sunt-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 |
| } |
| }, |
| "nisi-sunt_lsat-lr_cot": { |
| "task": "nisi-sunt_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "nisi-sunt-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 |
| } |
| }, |
| "nisi-sunt_lsat-rc_cot": { |
| "task": "nisi-sunt_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "nisi-sunt-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 |
| } |
| }, |
| "saepe-fuga_logiqa2_cot": { |
| "task": "saepe-fuga_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "saepe-fuga-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 |
| } |
| }, |
| "saepe-fuga_logiqa_cot": { |
| "task": "saepe-fuga_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "saepe-fuga-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 |
| } |
| }, |
| "saepe-fuga_lsat-ar_cot": { |
| "task": "saepe-fuga_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "saepe-fuga-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 |
| } |
| }, |
| "saepe-fuga_lsat-lr_cot": { |
| "task": "saepe-fuga_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "saepe-fuga-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 |
| } |
| }, |
| "saepe-fuga_lsat-rc_cot": { |
| "task": "saepe-fuga_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "saepe-fuga-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 |
| } |
| }, |
| "veritatis-velit_logiqa2_cot": { |
| "task": "veritatis-velit_logiqa2_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "veritatis-velit-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 |
| } |
| }, |
| "veritatis-velit_logiqa_cot": { |
| "task": "veritatis-velit_logiqa_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "veritatis-velit-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 |
| } |
| }, |
| "veritatis-velit_lsat-ar_cot": { |
| "task": "veritatis-velit_lsat-ar_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "veritatis-velit-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 |
| } |
| }, |
| "veritatis-velit_lsat-lr_cot": { |
| "task": "veritatis-velit_lsat-lr_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "veritatis-velit-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 |
| } |
| }, |
| "veritatis-velit_lsat-rc_cot": { |
| "task": "veritatis-velit_lsat-rc_cot", |
| "group": "logikon-bench", |
| "dataset_path": "logikon/cot-eval-traces", |
| "dataset_kwargs": { |
| "data_files": { |
| "test": "veritatis-velit-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": { |
| "eum-saepe_logiqa2_cot": 0.0, |
| "eum-saepe_logiqa_cot": 0.0, |
| "eum-saepe_lsat-ar_cot": 0.0, |
| "eum-saepe_lsat-lr_cot": 0.0, |
| "eum-saepe_lsat-rc_cot": 0.0, |
| "iste-molestias_logiqa2_cot": 0.0, |
| "iste-molestias_logiqa_cot": 0.0, |
| "iste-molestias_lsat-ar_cot": 0.0, |
| "iste-molestias_lsat-lr_cot": 0.0, |
| "iste-molestias_lsat-rc_cot": 0.0, |
| "laboriosam-molestiae_logiqa2_cot": 0.0, |
| "laboriosam-molestiae_logiqa_cot": 0.0, |
| "laboriosam-molestiae_lsat-ar_cot": 0.0, |
| "laboriosam-molestiae_lsat-lr_cot": 0.0, |
| "laboriosam-molestiae_lsat-rc_cot": 0.0, |
| "nisi-sunt_logiqa2_cot": 0.0, |
| "nisi-sunt_logiqa_cot": 0.0, |
| "nisi-sunt_lsat-ar_cot": 0.0, |
| "nisi-sunt_lsat-lr_cot": 0.0, |
| "nisi-sunt_lsat-rc_cot": 0.0, |
| "saepe-fuga_logiqa2_cot": 0.0, |
| "saepe-fuga_logiqa_cot": 0.0, |
| "saepe-fuga_lsat-ar_cot": 0.0, |
| "saepe-fuga_lsat-lr_cot": 0.0, |
| "saepe-fuga_lsat-rc_cot": 0.0, |
| "veritatis-velit_logiqa2_cot": 0.0, |
| "veritatis-velit_logiqa_cot": 0.0, |
| "veritatis-velit_lsat-ar_cot": 0.0, |
| "veritatis-velit_lsat-lr_cot": 0.0, |
| "veritatis-velit_lsat-rc_cot": 0.0 |
| }, |
| "n-shot": { |
| "eum-saepe_logiqa2_cot": 0, |
| "eum-saepe_logiqa_cot": 0, |
| "eum-saepe_lsat-ar_cot": 0, |
| "eum-saepe_lsat-lr_cot": 0, |
| "eum-saepe_lsat-rc_cot": 0, |
| "iste-molestias_logiqa2_cot": 0, |
| "iste-molestias_logiqa_cot": 0, |
| "iste-molestias_lsat-ar_cot": 0, |
| "iste-molestias_lsat-lr_cot": 0, |
| "iste-molestias_lsat-rc_cot": 0, |
| "laboriosam-molestiae_logiqa2_cot": 0, |
| "laboriosam-molestiae_logiqa_cot": 0, |
| "laboriosam-molestiae_lsat-ar_cot": 0, |
| "laboriosam-molestiae_lsat-lr_cot": 0, |
| "laboriosam-molestiae_lsat-rc_cot": 0, |
| "nisi-sunt_logiqa2_cot": 0, |
| "nisi-sunt_logiqa_cot": 0, |
| "nisi-sunt_lsat-ar_cot": 0, |
| "nisi-sunt_lsat-lr_cot": 0, |
| "nisi-sunt_lsat-rc_cot": 0, |
| "saepe-fuga_logiqa2_cot": 0, |
| "saepe-fuga_logiqa_cot": 0, |
| "saepe-fuga_lsat-ar_cot": 0, |
| "saepe-fuga_lsat-lr_cot": 0, |
| "saepe-fuga_lsat-rc_cot": 0, |
| "veritatis-velit_logiqa2_cot": 0, |
| "veritatis-velit_logiqa_cot": 0, |
| "veritatis-velit_lsat-ar_cot": 0, |
| "veritatis-velit_lsat-lr_cot": 0, |
| "veritatis-velit_lsat-rc_cot": 0 |
| }, |
| "config": { |
| "model": "vllm", |
| "model_args": "pretrained=Deci/DeciLM-7B,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": "5044cf9" |
| } |
