results/GPT-4o.json

{"passkey": {"4k": 1.0, "8k": 1.0, "16k": 1.0, "32k": 1.0, "64k": 1.0, "128k": 1.0, "dataset_total_score": 1.0}, "matreshka_yes_no": {"4k": 0.8, "8k": 0.6, "16k": 1.0, "32k": 0.8, "64k": 0.7, "128k": 0.9, "dataset_total_score": 0.8000000000000002}, "matreshka_names": {"4k": 0.6, "8k": 0.6, "16k": 0.5, "32k": 0.4, "64k": 0.5, "128k": 0.5, "dataset_total_score": 0.5166666666666667}, "passkey_with_librusec": {"4k": 1.0, "8k": 1.0, "16k": 1.0, "32k": 1.0, "64k": 1.0, "128k": 1.0, "dataset_total_score": 1.0}, "librusec_history": {"8k": 1.0, "16k": 1.0, "32k": 1.0, "64k": 0.9, "dataset_total_score": 0.975}, "ru_gsm100": {"16k": 1.0, "dataset_total_score": 1.0}, "ru_sci_passage_count": {"4k": 1.0, "8k": 0.5, "16k": 0.3, "32k": 0.0, "64k": 0.2, "128k": 0.1, "dataset_total_score": 0.35000000000000003}, "ru_2wikimultihopqa": {"8k": 0.8, "16k": 0.8, "32k": 0.7, "dataset_total_score": 0.7666666666666666}, "long_context_multiq": {"4k": 0.3, "8k": 1.0, "16k": 0.7, "32k": 0.0, "64k": 0.1, "128k": 0.1, "dataset_total_score": 0.3666666666666667}, "ru_sci_abstract_retrieval": {"4k": 0.99, "8k": 0.9541666666666668, "16k": 0.9254479578392623, "32k": 0.9562564463343153, "64k": 0.590978869808793, "128k": 0.19764315322255238, "dataset_total_score": 0.7690821823119315}, "ru_trec": {"4k": 0.6, "8k": 0.8, "16k": 0.9, "32k": 0.7, "dataset_total_score": 0.75}, "ru_sci_fi": {"32k": 0.6, "64k": 0.9, "dataset_total_score": 0.75}, "librusec_mhqa": {"8k": 0.5, "dataset_total_score": 0.5}, "ru_babilong_qa1": {"4k": 0.9, "8k": 0.8, "16k": 0.7, "32k": 0.9, "64k": 0.8, "128k": 0.6, "dataset_total_score": 0.7833333333333333}, "ru_babilong_qa2": {"4k": 0.4, "8k": 0.3, "16k": 0.4, "32k": 0.4, "64k": 0.5, "128k": 0.2, "dataset_total_score": 0.3666666666666667}, "ru_babilong_qa3": {"4k": 0.2, "8k": 0.3, "16k": 0.1, "32k": 0.2, "64k": 0.2, "128k": 0.2866666666666667, "dataset_total_score": 0.21444444444444444}, "ru_babilong_qa4": {"4k": 0.8800000000000001, "8k": 0.8, "16k": 0.8, "32k": 0.5714285714285715, "64k": 0.8857142857142858, "128k": 0.8, "dataset_total_score": 0.7895238095238096}, "ru_babilong_qa5": {"4k": 0.8666666666666666, "8k": 0.8666666666666666, "16k": 0.9333333333333333, "32k": 0.9666666666666666, "64k": 0.8666666666666668, "128k": 0.9, "dataset_total_score": 0.9}, "ru_quality": {"8k": 0.8, "16k": 0.8666666666666668, "dataset_total_score": 0.8333333333333335}, "ru_tpo": {"8k": 1.0, "dataset_total_score": 1.0}, "ru_qasper": {"8k": 0.2865100250626566, "16k": 0.3184757236227824, "32k": 0.3465384615384616, "dataset_total_score": 0.3171747367413002}, "total_score": 0.70231230982642}