{ "results": { "agieval": { "acc,none": 0.663159167876149, "acc_stderr,none": 0.004392357670686218, "alias": "agieval" }, "agieval_aqua_rat": { "alias": " - agieval_aqua_rat", "acc,none": 0.46062992125984253, "acc_stderr,none": 0.031337131298568036, "acc_norm,none": 0.43700787401574803, "acc_norm_stderr,none": 0.031184266331855014 }, "agieval_gaokao_biology": { "alias": " - agieval_gaokao_biology", "acc,none": 0.8857142857142857, "acc_stderr,none": 0.02200744652095776, "acc_norm,none": 0.861904761904762, "acc_norm_stderr,none": 0.02386414332035886 }, "agieval_gaokao_chemistry": { "alias": " - agieval_gaokao_chemistry", "acc,none": 0.6908212560386473, "acc_stderr,none": 0.03219986494000449, "acc_norm,none": 0.5797101449275363, "acc_norm_stderr,none": 0.03439111795440137 }, "agieval_gaokao_chinese": { "alias": " - agieval_gaokao_chinese", "acc,none": 0.8536585365853658, "acc_stderr,none": 0.0225809780432329, "acc_norm,none": 0.8333333333333334, "acc_norm_stderr,none": 0.023809523809523836 }, "agieval_gaokao_english": { "alias": " - agieval_gaokao_english", "acc,none": 0.8790849673202614, "acc_stderr,none": 0.018668338020084146, "acc_norm,none": 0.9084967320261438, "acc_norm_stderr,none": 0.01650935352607882 }, "agieval_gaokao_geography": { "alias": " - agieval_gaokao_geography", "acc,none": 0.8442211055276382, "acc_stderr,none": 0.025772100500124857, "acc_norm,none": 0.8391959798994975, "acc_norm_stderr,none": 0.026106433978056186 }, "agieval_gaokao_history": { "alias": " - agieval_gaokao_history", "acc,none": 0.8893617021276595, "acc_stderr,none": 0.020506145099008433, "acc_norm,none": 0.9106382978723404, "acc_norm_stderr,none": 0.01864836423253194 }, "agieval_gaokao_mathcloze": { "alias": " - agieval_gaokao_mathcloze", "acc,none": 0.0423728813559322, "acc_stderr,none": 0.018622984668462274 }, "agieval_gaokao_mathqa": { "alias": " - agieval_gaokao_mathqa", "acc,none": 0.5783475783475783, "acc_stderr,none": 0.02639597680205238, "acc_norm,none": 0.5527065527065527, "acc_norm_stderr,none": 0.026577220068633042 }, "agieval_gaokao_physics": { "alias": " - agieval_gaokao_physics", "acc,none": 0.755, "acc_stderr,none": 0.03048807329211421, "acc_norm,none": 0.705, "acc_norm_stderr,none": 0.032328014206142675 }, "agieval_jec_qa_ca": { "alias": " - agieval_jec_qa_ca", "acc,none": 0.8128128128128128, "acc_stderr,none": 0.012347187948703799, "acc_norm,none": 0.7837837837837838, "acc_norm_stderr,none": 0.01303097758477811 }, "agieval_jec_qa_kd": { "alias": " - agieval_jec_qa_kd", "acc,none": 0.881, "acc_stderr,none": 0.010244215145336667, "acc_norm,none": 0.851, "acc_norm_stderr,none": 0.011266140684632171 }, "agieval_logiqa_en": { "alias": " - agieval_logiqa_en", "acc,none": 0.6036866359447005, "acc_stderr,none": 0.019185294108788765, "acc_norm,none": 0.6098310291858678, "acc_norm_stderr,none": 0.019132619951195386 }, "agieval_logiqa_zh": { "alias": " - agieval_logiqa_zh", "acc,none": 0.6820276497695853, "acc_stderr,none": 0.01826581231613446, "acc_norm,none": 0.6574500768049155, "acc_norm_stderr,none": 0.01861386882920801 }, "agieval_lsat_ar": { "alias": " - agieval_lsat_ar", "acc,none": 0.3130434782608696, "acc_stderr,none": 0.03064426536742552, "acc_norm,none": 0.2956521739130435, "acc_norm_stderr,none": 0.030155489768916174 }, "agieval_lsat_lr": { "alias": " - agieval_lsat_lr", "acc,none": 0.7901960784313725, "acc_stderr,none": 0.01804742911247608, "acc_norm,none": 0.7843137254901961, "acc_norm_stderr,none": 0.018230445049830818 }, "agieval_lsat_rc": { "alias": " - agieval_lsat_rc", "acc,none": 0.8364312267657993, "acc_stderr,none": 0.02259424950424165, "acc_norm,none": 0.8327137546468402, "acc_norm_stderr,none": 0.022798726518245306 }, "agieval_math": { "alias": " - agieval_math", "acc,none": 0.127, "acc_stderr,none": 0.010534798620855755 }, "agieval_sat_en": { "alias": " - agieval_sat_en", "acc,none": 0.9174757281553398, "acc_stderr,none": 0.019218133764014527, "acc_norm,none": 0.9174757281553398, "acc_norm_stderr,none": 0.019218133764014527 }, "agieval_sat_en_without_passage": { "alias": " - agieval_sat_en_without_passage", "acc,none": 0.5922330097087378, "acc_stderr,none": 0.03432222290260261, "acc_norm,none": 0.5922330097087378, "acc_norm_stderr,none": 0.03432222290260261 }, "agieval_sat_math": { "alias": " - agieval_sat_math", "acc,none": 0.7, "acc_stderr,none": 0.030966176864266656, "acc_norm,none": 0.6727272727272727, "acc_norm_stderr,none": 0.03170679667686021 } }, "groups": { "agieval": { "acc,none": 0.663159167876149, "acc_stderr,none": 0.004392357670686218, "alias": "agieval" } }, "group_subtasks": { "agieval": [ "agieval_gaokao_biology", "agieval_gaokao_chemistry", "agieval_gaokao_chinese", "agieval_gaokao_geography", "agieval_gaokao_history", "agieval_gaokao_mathcloze", "agieval_gaokao_mathqa", "agieval_gaokao_physics", "agieval_jec_qa_ca", "agieval_jec_qa_kd", "agieval_logiqa_zh", "agieval_aqua_rat", "agieval_gaokao_english", "agieval_logiqa_en", "agieval_lsat_ar", "agieval_lsat_lr", "agieval_lsat_rc", "agieval_math", "agieval_sat_en_without_passage", "agieval_sat_en", "agieval_sat_math" ] }, "configs": { "agieval_aqua_rat": { "task": "agieval_aqua_rat", "dataset_path": "hails/agieval-aqua-rat", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_biology": { "task": "agieval_gaokao_biology", "dataset_path": "hails/agieval-gaokao-biology", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_chemistry": { "task": "agieval_gaokao_chemistry", "dataset_path": "hails/agieval-gaokao-chemistry", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_chinese": { "task": "agieval_gaokao_chinese", "dataset_path": "hails/agieval-gaokao-chinese", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_english": { "task": "agieval_gaokao_english", "dataset_path": "hails/agieval-gaokao-english", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_geography": { "task": "agieval_gaokao_geography", "dataset_path": "hails/agieval-gaokao-geography", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_history": { "task": "agieval_gaokao_history", "dataset_path": "hails/agieval-gaokao-history", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_mathcloze": { "task": "agieval_gaokao_mathcloze", "dataset_path": "hails/agieval-gaokao-mathcloze", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{answer}}", "process_results": "def process_results(doc: dict, results: List[str]) -> Dict[str, int]:\n candidate = results[0]\n\n gold = doc[\"answer\"]\n\n if not gold:\n print(doc, candidate, gold)\n if is_equiv(candidate, gold):\n retval = 1\n else:\n retval = 0\n\n results = {\n \"acc\": retval,\n }\n return results\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true } ], "output_type": "generate_until", "generation_kwargs": { "max_gen_toks": 32, "do_sample": false, "temperature": 0.0, "until": [ "Q:" ] }, "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_mathqa": { "task": "agieval_gaokao_mathqa", "dataset_path": "hails/agieval-gaokao-mathqa", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_gaokao_physics": { "task": "agieval_gaokao_physics", "dataset_path": "hails/agieval-gaokao-physics", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_jec_qa_ca": { "task": "agieval_jec_qa_ca", "dataset_path": "hails/agieval-jec-qa-ca", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_jec_qa_kd": { "task": "agieval_jec_qa_kd", "dataset_path": "hails/agieval-jec-qa-kd", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_logiqa_en": { "task": "agieval_logiqa_en", "dataset_path": "hails/agieval-logiqa-en", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_logiqa_zh": { "task": "agieval_logiqa_zh", "dataset_path": "hails/agieval-logiqa-zh", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_lsat_ar": { "task": "agieval_lsat_ar", "dataset_path": "hails/agieval-lsat-ar", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_lsat_lr": { "task": "agieval_lsat_lr", "dataset_path": "hails/agieval-lsat-lr", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_lsat_rc": { "task": "agieval_lsat_rc", "dataset_path": "hails/agieval-lsat-rc", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_math": { "task": "agieval_math", "dataset_path": "hails/agieval-math", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{answer}}", "process_results": "def process_results(doc: dict, results: List[str]) -> Dict[str, int]:\n candidate = results[0]\n\n gold = doc[\"answer\"]\n\n if not gold:\n print(doc, candidate, gold)\n if is_equiv(candidate, gold):\n retval = 1\n else:\n retval = 0\n\n results = {\n \"acc\": retval,\n }\n return results\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true } ], "output_type": "generate_until", "generation_kwargs": { "max_gen_toks": 32, "do_sample": false, "temperature": 0.0, "until": [ "Q:" ] }, "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_sat_en": { "task": "agieval_sat_en", "dataset_path": "hails/agieval-sat-en", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_sat_en_without_passage": { "task": "agieval_sat_en_without_passage", "dataset_path": "hails/agieval-sat-en-without-passage", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } }, "agieval_sat_math": { "task": "agieval_sat_math", "dataset_path": "hails/agieval-sat-math", "test_split": "test", "doc_to_text": "{{query}}", "doc_to_target": "{{gold}}", "doc_to_choice": "{{choices}}", "process_results": "def process_results_mcqa(doc, results):\n results = [result[0] for result in results]\n\n gold = doc[\"gold\"]\n\n acc = 1.0 if int(np.argmax(results)) in gold else 0.0\n completion_len = np.array([float(len(i)) for i in doc[\"choices\"]])\n acc_norm = 1.0 if int(np.argmax(results / completion_len)) in gold else 0.0\n\n return {\n \"acc\": acc,\n \"acc_norm\": acc_norm,\n }\n", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc", "aggregation": "mean", "higher_is_better": true }, { "metric": "acc_norm", "aggregation": "mean", "higher_is_better": true } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": false, "metadata": { "version": 1.0 } } }, "versions": { "agieval": 0.0, "agieval_aqua_rat": 1.0, "agieval_gaokao_biology": 1.0, "agieval_gaokao_chemistry": 1.0, "agieval_gaokao_chinese": 1.0, "agieval_gaokao_english": 1.0, "agieval_gaokao_geography": 1.0, "agieval_gaokao_history": 1.0, "agieval_gaokao_mathcloze": 1.0, "agieval_gaokao_mathqa": 1.0, "agieval_gaokao_physics": 1.0, "agieval_jec_qa_ca": 1.0, "agieval_jec_qa_kd": 1.0, "agieval_logiqa_en": 1.0, "agieval_logiqa_zh": 1.0, "agieval_lsat_ar": 1.0, "agieval_lsat_lr": 1.0, "agieval_lsat_rc": 1.0, "agieval_math": 1.0, "agieval_sat_en": 1.0, "agieval_sat_en_without_passage": 1.0, "agieval_sat_math": 1.0 }, "n-shot": { "agieval_aqua_rat": 0, "agieval_gaokao_biology": 0, "agieval_gaokao_chemistry": 0, "agieval_gaokao_chinese": 0, "agieval_gaokao_english": 0, "agieval_gaokao_geography": 0, "agieval_gaokao_history": 0, "agieval_gaokao_mathcloze": 0, "agieval_gaokao_mathqa": 0, "agieval_gaokao_physics": 0, "agieval_jec_qa_ca": 0, "agieval_jec_qa_kd": 0, "agieval_logiqa_en": 0, "agieval_logiqa_zh": 0, "agieval_lsat_ar": 0, "agieval_lsat_lr": 0, "agieval_lsat_rc": 0, "agieval_math": 0, "agieval_sat_en": 0, "agieval_sat_en_without_passage": 0, "agieval_sat_math": 0 }, "higher_is_better": { "agieval": { "acc": true, "acc_norm": true }, "agieval_aqua_rat": { "acc": true, "acc_norm": true }, "agieval_gaokao_biology": { "acc": true, "acc_norm": true }, "agieval_gaokao_chemistry": { "acc": true, "acc_norm": true }, "agieval_gaokao_chinese": { "acc": true, "acc_norm": true }, "agieval_gaokao_english": { "acc": true, "acc_norm": true }, "agieval_gaokao_geography": { "acc": true, "acc_norm": true }, "agieval_gaokao_history": { "acc": true, "acc_norm": true }, "agieval_gaokao_mathcloze": { "acc": true }, "agieval_gaokao_mathqa": { "acc": true, "acc_norm": true }, "agieval_gaokao_physics": { "acc": true, "acc_norm": true }, "agieval_jec_qa_ca": { "acc": true, "acc_norm": true }, "agieval_jec_qa_kd": { "acc": true, "acc_norm": true }, "agieval_logiqa_en": { "acc": true, "acc_norm": true }, "agieval_logiqa_zh": { "acc": true, "acc_norm": true }, "agieval_lsat_ar": { "acc": true, "acc_norm": true }, "agieval_lsat_lr": { "acc": true, "acc_norm": true }, "agieval_lsat_rc": { "acc": true, "acc_norm": true }, "agieval_math": { "acc": true }, "agieval_sat_en": { "acc": true, "acc_norm": true }, "agieval_sat_en_without_passage": { "acc": true, "acc_norm": true }, "agieval_sat_math": { "acc": true, "acc_norm": true } }, "n-samples": { "agieval_gaokao_biology": { "original": 210, "effective": 210 }, "agieval_gaokao_chemistry": { "original": 207, "effective": 207 }, "agieval_gaokao_chinese": { "original": 246, "effective": 246 }, "agieval_gaokao_geography": { "original": 199, "effective": 199 }, "agieval_gaokao_history": { "original": 235, "effective": 235 }, "agieval_gaokao_mathcloze": { "original": 118, "effective": 118 }, "agieval_gaokao_mathqa": { "original": 351, "effective": 351 }, "agieval_gaokao_physics": { "original": 200, "effective": 200 }, "agieval_jec_qa_ca": { "original": 999, "effective": 999 }, "agieval_jec_qa_kd": { "original": 1000, "effective": 1000 }, "agieval_logiqa_zh": { "original": 651, "effective": 651 }, "agieval_aqua_rat": { "original": 254, "effective": 254 }, "agieval_gaokao_english": { "original": 306, "effective": 306 }, "agieval_logiqa_en": { "original": 651, "effective": 651 }, "agieval_lsat_ar": { "original": 230, "effective": 230 }, "agieval_lsat_lr": { "original": 510, "effective": 510 }, "agieval_lsat_rc": { "original": 269, "effective": 269 }, "agieval_math": { "original": 1000, "effective": 1000 }, "agieval_sat_en_without_passage": { "original": 206, "effective": 206 }, "agieval_sat_en": { "original": 206, "effective": 206 }, "agieval_sat_math": { "original": 220, "effective": 220 } }, "config": { "model": "hf", "model_args": "pretrained=Qwen/Qwen2.5-14B-Instruct,parallelize=True,trust_remote_code=True,cache_dir=/tmp", "model_num_parameters": 14770033664, "model_dtype": "torch.bfloat16", "model_revision": "main", "model_sha": "cf98f3b3bbb457ad9e2bb7baf9a0125b6b88caa8", "batch_size": "auto", "batch_sizes": [ 16 ], "device": null, "use_cache": null, "limit": null, "bootstrap_iters": 100000, "gen_kwargs": null, "random_seed": 0, "numpy_seed": 1234, "torch_seed": 1234, "fewshot_seed": 1234 }, "git_hash": "8e1bd48d", "date": 1736182802.75474, "pretty_env_info": "PyTorch version: 2.4.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: Ubuntu 22.04.3 LTS (x86_64)\nGCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0\nClang version: Could not collect\nCMake version: version 3.27.9\nLibc version: glibc-2.35\n\nPython version: 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0] (64-bit runtime)\nPython platform: Linux-5.15.0-1064-azure-x86_64-with-glibc2.35\nIs CUDA available: True\nCUDA runtime version: 12.3.107\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: \nGPU 0: NVIDIA A100 80GB PCIe\nGPU 1: NVIDIA A100 80GB PCIe\nGPU 2: NVIDIA A100 80GB PCIe\nGPU 3: NVIDIA A100 80GB PCIe\n\nNvidia driver version: 535.161.08\ncuDNN version: Probably one of the following:\n/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.7\n/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.7\n/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.7\n/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.7\n/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.7\n/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.7\n/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.7\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nAddress sizes: 48 bits physical, 48 bits virtual\nByte Order: Little Endian\nCPU(s): 96\nOn-line CPU(s) list: 0-95\nVendor ID: AuthenticAMD\nModel name: AMD EPYC 7V13 64-Core Processor\nCPU family: 25\nModel: 1\nThread(s) per core: 1\nCore(s) per socket: 48\nSocket(s): 2\nStepping: 1\nBogoMIPS: 4890.87\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl tsc_reliable nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw topoext perfctr_core invpcid_single vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr rdpru arat umip vaes vpclmulqdq rdpid fsrm\nHypervisor vendor: Microsoft\nVirtualization type: full\nL1d cache: 3 MiB (96 instances)\nL1i cache: 3 MiB (96 instances)\nL2 cache: 48 MiB (96 instances)\nL3 cache: 384 MiB (12 instances)\nNUMA node(s): 4\nNUMA node0 CPU(s): 0-23\nNUMA node1 CPU(s): 24-47\nNUMA node2 CPU(s): 48-71\nNUMA node3 CPU(s): 72-95\nVulnerability Gather data sampling: Not affected\nVulnerability Itlb multihit: Not affected\nVulnerability L1tf: Not affected\nVulnerability Mds: Not affected\nVulnerability Meltdown: Not affected\nVulnerability Mmio stale data: Not affected\nVulnerability Retbleed: Not affected\nVulnerability Spec rstack overflow: Mitigation; safe RET, no microcode\nVulnerability Spec store bypass: Vulnerable\nVulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVulnerability Spectre v2: Mitigation; Retpolines; STIBP disabled; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected\nVulnerability Srbds: Not affected\nVulnerability Tsx async abort: Not affected\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] onnx==1.15.0rc2\n[pip3] open_clip_torch==2.26.1\n[pip3] optree==0.10.0\n[pip3] pytorch-lightning==2.0.7\n[pip3] pytorch-quantization==2.1.2\n[pip3] torch==2.4.0\n[pip3] torch-tensorrt==2.2.0a0\n[pip3] torchdata==0.7.0a0\n[pip3] torchdiffeq==0.2.4\n[pip3] torchmetrics==1.4.1\n[pip3] torchsde==0.2.6\n[pip3] torchtext==0.17.0a0\n[pip3] torchvision==0.19.0\n[pip3] triton==3.0.0\n[conda] Could not collect", "transformers_version": "4.44.0", "upper_git_hash": null, "tokenizer_pad_token": [ "<|endoftext|>", "151643" ], "tokenizer_eos_token": [ "<|im_end|>", "151645" ], "tokenizer_bos_token": [ null, "None" ], "eot_token_id": 151645, "max_length": 32768, "task_hashes": {}, "model_source": "hf", "model_name": "Qwen/Qwen2.5-14B-Instruct", "model_name_sanitized": "Qwen__Qwen2.5-14B-Instruct", "system_instruction": null, "system_instruction_sha": null, "fewshot_as_multiturn": false, "chat_template": null, "chat_template_sha": null, "start_time": 39100.313463795, "end_time": 43387.004770483, "total_evaluation_time_seconds": "4286.691306688001" }