--- tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:124788 - loss:GISTEmbedLoss base_model: BAAI/bge-small-en-v1.5 widget: - source_sentence: 其他机械、设备和有形货物租赁服务代表 sentences: - 其他机械和设备租赁服务工作人员 - 电子和电信设备及零部件物流经理 - 工业主厨 - source_sentence: 公交车司机 sentences: - 表演灯光设计师 - 乙烯基地板安装工 - 国际巴士司机 - source_sentence: online communication manager sentences: - trades union official - social media manager - budget manager - source_sentence: Projektmanagerin sentences: - Projektmanager/Projektmanagerin - Category-Manager - Infanterist - source_sentence: Volksvertreter sentences: - Parlamentarier - Oberbürgermeister - Konsul pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - cosine_accuracy@1 - cosine_accuracy@20 - cosine_accuracy@50 - cosine_accuracy@100 - cosine_accuracy@150 - cosine_accuracy@200 - cosine_precision@1 - cosine_precision@20 - cosine_precision@50 - cosine_precision@100 - cosine_precision@150 - cosine_precision@200 - cosine_recall@1 - cosine_recall@20 - cosine_recall@50 - cosine_recall@100 - cosine_recall@150 - cosine_recall@200 - cosine_ndcg@1 - cosine_ndcg@20 - cosine_ndcg@50 - cosine_ndcg@100 - cosine_ndcg@150 - cosine_ndcg@200 - cosine_mrr@1 - cosine_mrr@20 - cosine_mrr@50 - cosine_mrr@100 - cosine_mrr@150 - cosine_mrr@200 - cosine_map@1 - cosine_map@20 - cosine_map@50 - cosine_map@100 - cosine_map@150 - cosine_map@200 - cosine_map@500 model-index: - name: SentenceTransformer based on BAAI/bge-small-en-v1.5 results: - task: type: information-retrieval name: Information Retrieval dataset: name: full en type: full_en metrics: - type: cosine_accuracy@1 value: 0.6571428571428571 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 0.9904761904761905 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 0.9904761904761905 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 0.9904761904761905 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 0.9904761904761905 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 0.9904761904761905 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.6571428571428571 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.5057142857142858 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.3085714285714286 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.18685714285714286 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.13263492063492063 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.10261904761904762 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.0680237860830842 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.539814746481506 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.7281788406466259 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.8369695734692713 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.8797734908498225 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.9040821090543185 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.6571428571428571 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.6871892352981543 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.7057435134474674 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.7611594394123498 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.7798336860589586 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.7894131768304745 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.6571428571428571 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.8095238095238095 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.8095238095238095 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.8095238095238095 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.8095238095238095 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.8095238095238095 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.6571428571428571 name: Cosine Map@1 - type: cosine_map@20 value: 0.5458017299619246 name: Cosine Map@20 - type: cosine_map@50 value: 0.5350568967293148 name: Cosine Map@50 - type: cosine_map@100 value: 0.5681338314009312 name: Cosine Map@100 - type: cosine_map@150 value: 0.5758337072896192 name: Cosine Map@150 - type: cosine_map@200 value: 0.5788774324789392 name: Cosine Map@200 - type: cosine_map@500 value: 0.5832951333498196 name: Cosine Map@500 - task: type: information-retrieval name: Information Retrieval dataset: name: full es type: full_es metrics: - type: cosine_accuracy@1 value: 0.12432432432432433 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 1.0 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 1.0 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 1.0 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 1.0 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 1.0 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.12432432432432433 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.49054054054054047 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.3163243243243243 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.19794594594594597 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.1476036036036036 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.11764864864864866 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.003111544931768446 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.3226281360780687 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.460233186838451 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.541868009988165 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.5852603494024129 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.6129186722388266 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.12432432432432433 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.539216162208845 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.4996835226060237 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.5137905428062277 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.5360687473286022 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.5499397431446335 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.12432432432432433 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.54987987987988 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.54987987987988 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.54987987987988 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.54987987987988 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.54987987987988 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.12432432432432433 name: Cosine Map@1 - type: cosine_map@20 value: 0.4038320848933988 name: Cosine Map@20 - type: cosine_map@50 value: 0.34180322548979025 name: Cosine Map@50 - type: cosine_map@100 value: 0.33919520921684143 name: Cosine Map@100 - type: cosine_map@150 value: 0.34836248453935964 name: Cosine Map@150 - type: cosine_map@200 value: 0.3534621616433109 name: Cosine Map@200 - type: cosine_map@500 value: 0.36317607366795296 name: Cosine Map@500 - task: type: information-retrieval name: Information Retrieval dataset: name: full de type: full_de metrics: - type: cosine_accuracy@1 value: 0.2955665024630542 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 0.9211822660098522 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 0.9655172413793104 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 0.9753694581280788 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 0.9852216748768473 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 0.9852216748768473 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.2955665024630542 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.4273399014778325 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.2806896551724138 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.17970443349753695 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.13536945812807882 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.11017241379310345 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.01108543831680986 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.2624274710898933 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.3816838951440551 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.466648189283344 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.5156350937800631 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.5501200676805387 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.2955665024630542 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.4612630794197783 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.4218308857158181 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.43630012299200127 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.4611671590034933 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.47859447889613466 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.2955665024630542 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.4902263308667246 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.49157690165668283 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.491727421634789 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.49180457899033725 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.49180457899033725 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.2955665024630542 name: Cosine Map@1 - type: cosine_map@20 value: 0.32403320158333687 name: Cosine Map@20 - type: cosine_map@50 value: 0.263929499089501 name: Cosine Map@50 - type: cosine_map@100 value: 0.2571656148623807 name: Cosine Map@100 - type: cosine_map@150 value: 0.2662223711722934 name: Cosine Map@150 - type: cosine_map@200 value: 0.2716128199441605 name: Cosine Map@200 - type: cosine_map@500 value: 0.2826472699200774 name: Cosine Map@500 - task: type: information-retrieval name: Information Retrieval dataset: name: full zh type: full_zh metrics: - type: cosine_accuracy@1 value: 0.30097087378640774 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 0.6990291262135923 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 0.8446601941747572 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 0.883495145631068 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 0.941747572815534 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 0.941747572815534 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.30097087378640774 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.16504854368932043 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.0932038834951456 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.05825242718446601 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.04601941747572816 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.038834951456310676 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.02386287516726942 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.1765726567124355 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.23652817418030478 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.2843496525793469 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.3277225727000478 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.3596812771438104 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.30097087378640774 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.23858136398044388 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.23394297561992747 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.25614039865630267 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.2739893276218014 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.2869614126642384 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.30097087378640774 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.4190465134339312 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.4240176433261117 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.4245497320288698 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.4250484654944024 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.4250484654944024 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.30097087378640774 name: Cosine Map@1 - type: cosine_map@20 value: 0.1429307444790631 name: Cosine Map@20 - type: cosine_map@50 value: 0.12440485302105762 name: Cosine Map@50 - type: cosine_map@100 value: 0.12975707309312245 name: Cosine Map@100 - type: cosine_map@150 value: 0.1324151603536725 name: Cosine Map@150 - type: cosine_map@200 value: 0.13400586423473068 name: Cosine Map@200 - type: cosine_map@500 value: 0.13792557615284717 name: Cosine Map@500 - task: type: information-retrieval name: Information Retrieval dataset: name: mix es type: mix_es metrics: - type: cosine_accuracy@1 value: 0.40561622464898595 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 0.7555902236089443 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 0.8465938637545501 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 0.9053562142485699 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 0.9391575663026521 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 0.9521580863234529 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.40561622464898595 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.08801352054082164 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.0414560582423297 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.02274050962038482 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.01580863234529381 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.012181487259490382 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.15422578807914222 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.6504234455092489 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.7645209617908526 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.8407386771661344 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.8772614714112373 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.9011477601961221 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.40561622464898595 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.5052527283785204 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.5363668402911114 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.5529351239282845 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.5600105253735999 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.5643581500059056 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.40561622464898595 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.49193080862589195 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.49497073018920573 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.4958131715411002 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.4960911044064609 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.49616673564150066 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.40561622464898595 name: Cosine Map@1 - type: cosine_map@20 value: 0.42113760176243775 name: Cosine Map@20 - type: cosine_map@50 value: 0.4288029559186059 name: Cosine Map@50 - type: cosine_map@100 value: 0.4309354633228117 name: Cosine Map@100 - type: cosine_map@150 value: 0.43151233276792966 name: Cosine Map@150 - type: cosine_map@200 value: 0.4318026647046382 name: Cosine Map@200 - type: cosine_map@500 value: 0.4322791851958394 name: Cosine Map@500 - task: type: information-retrieval name: Information Retrieval dataset: name: mix de type: mix_de metrics: - type: cosine_accuracy@1 value: 0.28965158606344255 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 0.6453458138325533 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 0.7514300572022881 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 0.8424336973478939 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 0.8840353614144566 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 0.9131565262610505 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.28965158606344255 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.07230889235569424 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.035559022360894435 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.02045241809672387 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.014543248396602529 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.011443057722308895 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.10873634945397814 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.5285491419656786 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.6476772404229503 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.7456058242329694 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.7947564569249437 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.8333680013867221 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.28965158606344255 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.3869439859976832 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.419340853881164 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.4408718349726106 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.4505012036387736 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.4575667024678089 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.28965158606344255 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.37323361269727506 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.37671986743715985 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.37799876590389947 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.3783355727887503 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.37850251063580587 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.28965158606344255 name: Cosine Map@1 - type: cosine_map@20 value: 0.3048226727334811 name: Cosine Map@20 - type: cosine_map@50 value: 0.3123006074016708 name: Cosine Map@50 - type: cosine_map@100 value: 0.3150058759399239 name: Cosine Map@100 - type: cosine_map@150 value: 0.31578258870027714 name: Cosine Map@150 - type: cosine_map@200 value: 0.31623339353875296 name: Cosine Map@200 - type: cosine_map@500 value: 0.3169966358324732 name: Cosine Map@500 - task: type: information-retrieval name: Information Retrieval dataset: name: mix zh type: mix_zh metrics: - type: cosine_accuracy@1 value: 0.09551148225469729 name: Cosine Accuracy@1 - type: cosine_accuracy@20 value: 0.34812108559498955 name: Cosine Accuracy@20 - type: cosine_accuracy@50 value: 0.4932150313152401 name: Cosine Accuracy@50 - type: cosine_accuracy@100 value: 0.5970772442588727 name: Cosine Accuracy@100 - type: cosine_accuracy@150 value: 0.6623173277661796 name: Cosine Accuracy@150 - type: cosine_accuracy@200 value: 0.7035490605427975 name: Cosine Accuracy@200 - type: cosine_precision@1 value: 0.09551148225469729 name: Cosine Precision@1 - type: cosine_precision@20 value: 0.030480167014613778 name: Cosine Precision@20 - type: cosine_precision@50 value: 0.018402922755741128 name: Cosine Precision@50 - type: cosine_precision@100 value: 0.01144572025052192 name: Cosine Precision@100 - type: cosine_precision@150 value: 0.00861864996520529 name: Cosine Precision@150 - type: cosine_precision@200 value: 0.0070720250521920675 name: Cosine Precision@200 - type: cosine_recall@1 value: 0.03220250521920668 name: Cosine Recall@1 - type: cosine_recall@20 value: 0.20316259071478276 name: Cosine Recall@20 - type: cosine_recall@50 value: 0.3040399145044239 name: Cosine Recall@50 - type: cosine_recall@100 value: 0.37823300858269543 name: Cosine Recall@100 - type: cosine_recall@150 value: 0.4274327302250057 name: Cosine Recall@150 - type: cosine_recall@200 value: 0.467568429598701 name: Cosine Recall@200 - type: cosine_ndcg@1 value: 0.09551148225469729 name: Cosine Ndcg@1 - type: cosine_ndcg@20 value: 0.13449413074843178 name: Cosine Ndcg@20 - type: cosine_ndcg@50 value: 0.16281375869650408 name: Cosine Ndcg@50 - type: cosine_ndcg@100 value: 0.1798079117342116 name: Cosine Ndcg@100 - type: cosine_ndcg@150 value: 0.18976873702750593 name: Cosine Ndcg@150 - type: cosine_ndcg@200 value: 0.19737879358914515 name: Cosine Ndcg@200 - type: cosine_mrr@1 value: 0.09551148225469729 name: Cosine Mrr@1 - type: cosine_mrr@20 value: 0.148796848074501 name: Cosine Mrr@20 - type: cosine_mrr@50 value: 0.15342712486814447 name: Cosine Mrr@50 - type: cosine_mrr@100 value: 0.15489924735195337 name: Cosine Mrr@100 - type: cosine_mrr@150 value: 0.15543781318663716 name: Cosine Mrr@150 - type: cosine_mrr@200 value: 0.15567492413016254 name: Cosine Mrr@200 - type: cosine_map@1 value: 0.09551148225469729 name: Cosine Map@1 - type: cosine_map@20 value: 0.08472355170504958 name: Cosine Map@20 - type: cosine_map@50 value: 0.09038650929811239 name: Cosine Map@50 - type: cosine_map@100 value: 0.0922515804329945 name: Cosine Map@100 - type: cosine_map@150 value: 0.09296251507722719 name: Cosine Map@150 - type: cosine_map@200 value: 0.09340391507908284 name: Cosine Map@200 - type: cosine_map@500 value: 0.09434895514443004 name: Cosine Map@500 --- # SentenceTransformer based on BAAI/bge-small-en-v1.5 This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5) on the full_en, full_de, full_es, full_zh and mix datasets. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5) - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 384 dimensions - **Similarity Function:** Cosine Similarity - **Training Datasets:** - full_en - full_de - full_es - full_zh - mix ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) (2): Normalize() ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("sentence_transformers_model_id") # Run inference sentences = [ 'Volksvertreter', 'Parlamentarier', 'Oberbürgermeister', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 384] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` ## Evaluation ### Metrics #### Information Retrieval * Datasets: `full_en`, `full_es`, `full_de`, `full_zh`, `mix_es`, `mix_de` and `mix_zh` * Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) | Metric | full_en | full_es | full_de | full_zh | mix_es | mix_de | mix_zh | |:---------------------|:-----------|:-----------|:-----------|:----------|:-----------|:-----------|:-----------| | cosine_accuracy@1 | 0.6571 | 0.1243 | 0.2956 | 0.301 | 0.4056 | 0.2897 | 0.0955 | | cosine_accuracy@20 | 0.9905 | 1.0 | 0.9212 | 0.699 | 0.7556 | 0.6453 | 0.3481 | | cosine_accuracy@50 | 0.9905 | 1.0 | 0.9655 | 0.8447 | 0.8466 | 0.7514 | 0.4932 | | cosine_accuracy@100 | 0.9905 | 1.0 | 0.9754 | 0.8835 | 0.9054 | 0.8424 | 0.5971 | | cosine_accuracy@150 | 0.9905 | 1.0 | 0.9852 | 0.9417 | 0.9392 | 0.884 | 0.6623 | | cosine_accuracy@200 | 0.9905 | 1.0 | 0.9852 | 0.9417 | 0.9522 | 0.9132 | 0.7035 | | cosine_precision@1 | 0.6571 | 0.1243 | 0.2956 | 0.301 | 0.4056 | 0.2897 | 0.0955 | | cosine_precision@20 | 0.5057 | 0.4905 | 0.4273 | 0.165 | 0.088 | 0.0723 | 0.0305 | | cosine_precision@50 | 0.3086 | 0.3163 | 0.2807 | 0.0932 | 0.0415 | 0.0356 | 0.0184 | | cosine_precision@100 | 0.1869 | 0.1979 | 0.1797 | 0.0583 | 0.0227 | 0.0205 | 0.0114 | | cosine_precision@150 | 0.1326 | 0.1476 | 0.1354 | 0.046 | 0.0158 | 0.0145 | 0.0086 | | cosine_precision@200 | 0.1026 | 0.1176 | 0.1102 | 0.0388 | 0.0122 | 0.0114 | 0.0071 | | cosine_recall@1 | 0.068 | 0.0031 | 0.0111 | 0.0239 | 0.1542 | 0.1087 | 0.0322 | | cosine_recall@20 | 0.5398 | 0.3226 | 0.2624 | 0.1766 | 0.6504 | 0.5285 | 0.2032 | | cosine_recall@50 | 0.7282 | 0.4602 | 0.3817 | 0.2365 | 0.7645 | 0.6477 | 0.304 | | cosine_recall@100 | 0.837 | 0.5419 | 0.4666 | 0.2843 | 0.8407 | 0.7456 | 0.3782 | | cosine_recall@150 | 0.8798 | 0.5853 | 0.5156 | 0.3277 | 0.8773 | 0.7948 | 0.4274 | | cosine_recall@200 | 0.9041 | 0.6129 | 0.5501 | 0.3597 | 0.9011 | 0.8334 | 0.4676 | | cosine_ndcg@1 | 0.6571 | 0.1243 | 0.2956 | 0.301 | 0.4056 | 0.2897 | 0.0955 | | cosine_ndcg@20 | 0.6872 | 0.5392 | 0.4613 | 0.2386 | 0.5053 | 0.3869 | 0.1345 | | cosine_ndcg@50 | 0.7057 | 0.4997 | 0.4218 | 0.2339 | 0.5364 | 0.4193 | 0.1628 | | cosine_ndcg@100 | 0.7612 | 0.5138 | 0.4363 | 0.2561 | 0.5529 | 0.4409 | 0.1798 | | cosine_ndcg@150 | 0.7798 | 0.5361 | 0.4612 | 0.274 | 0.56 | 0.4505 | 0.1898 | | **cosine_ndcg@200** | **0.7894** | **0.5499** | **0.4786** | **0.287** | **0.5644** | **0.4576** | **0.1974** | | cosine_mrr@1 | 0.6571 | 0.1243 | 0.2956 | 0.301 | 0.4056 | 0.2897 | 0.0955 | | cosine_mrr@20 | 0.8095 | 0.5499 | 0.4902 | 0.419 | 0.4919 | 0.3732 | 0.1488 | | cosine_mrr@50 | 0.8095 | 0.5499 | 0.4916 | 0.424 | 0.495 | 0.3767 | 0.1534 | | cosine_mrr@100 | 0.8095 | 0.5499 | 0.4917 | 0.4245 | 0.4958 | 0.378 | 0.1549 | | cosine_mrr@150 | 0.8095 | 0.5499 | 0.4918 | 0.425 | 0.4961 | 0.3783 | 0.1554 | | cosine_mrr@200 | 0.8095 | 0.5499 | 0.4918 | 0.425 | 0.4962 | 0.3785 | 0.1557 | | cosine_map@1 | 0.6571 | 0.1243 | 0.2956 | 0.301 | 0.4056 | 0.2897 | 0.0955 | | cosine_map@20 | 0.5458 | 0.4038 | 0.324 | 0.1429 | 0.4211 | 0.3048 | 0.0847 | | cosine_map@50 | 0.5351 | 0.3418 | 0.2639 | 0.1244 | 0.4288 | 0.3123 | 0.0904 | | cosine_map@100 | 0.5681 | 0.3392 | 0.2572 | 0.1298 | 0.4309 | 0.315 | 0.0923 | | cosine_map@150 | 0.5758 | 0.3484 | 0.2662 | 0.1324 | 0.4315 | 0.3158 | 0.093 | | cosine_map@200 | 0.5789 | 0.3535 | 0.2716 | 0.134 | 0.4318 | 0.3162 | 0.0934 | | cosine_map@500 | 0.5833 | 0.3632 | 0.2826 | 0.1379 | 0.4323 | 0.317 | 0.0943 | ## Training Details ### Training Datasets
full_en #### full_en * Dataset: full_en * Size: 28,880 training samples * Columns: anchor and positive * Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:--------------------------------------------------------------------------------|:---------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:-----------------------------------------|:-----------------------------------------| | air commodore | flight lieutenant | | command and control officer | flight officer | | air commodore | command and control officer | * Loss: [GISTEmbedLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#gistembedloss) with these parameters: ```json {'guide': SentenceTransformer( (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) (2): Normalize() ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0} ```
full_de #### full_de * Dataset: full_de * Size: 23,023 training samples * Columns: anchor and positive * Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:----------------------------------|:-----------------------------------------------------| | Staffelkommandantin | Kommodore | | Luftwaffenoffizierin | Luftwaffenoffizier/Luftwaffenoffizierin | | Staffelkommandantin | Luftwaffenoffizierin | * Loss: [GISTEmbedLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#gistembedloss) with these parameters: ```json {'guide': SentenceTransformer( (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) (2): Normalize() ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0} ```
full_es #### full_es * Dataset: full_es * Size: 20,724 training samples * Columns: anchor and positive * Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:------------------------------------|:-------------------------------------------| | jefe de escuadrón | instructor | | comandante de aeronave | instructor de simulador | | instructor | oficial del Ejército del Aire | * Loss: [GISTEmbedLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#gistembedloss) with these parameters: ```json {'guide': SentenceTransformer( (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) (2): Normalize() ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0} ```
full_zh #### full_zh * Dataset: full_zh * Size: 30,401 training samples * Columns: anchor and positive * Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:------------------|:---------------------| | 技术总监 | 技术和运营总监 | | 技术总监 | 技术主管 | | 技术总监 | 技术艺术总监 | * Loss: [GISTEmbedLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#gistembedloss) with these parameters: ```json {'guide': SentenceTransformer( (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) (2): Normalize() ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0} ```
mix #### mix * Dataset: mix * Size: 21,760 training samples * Columns: anchor and positive * Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:------------------------------------------|:----------------------------------------------------------------| | technical manager | Technischer Direktor für Bühne, Film und Fernsehen | | head of technical | directora técnica | | head of technical department | 技术艺术总监 | * Loss: [GISTEmbedLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#gistembedloss) with these parameters: ```json {'guide': SentenceTransformer( (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) (2): Normalize() ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0} ```
### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 128 - `per_device_eval_batch_size`: 128 - `gradient_accumulation_steps`: 2 - `num_train_epochs`: 5 - `warmup_ratio`: 0.05 - `log_on_each_node`: False - `fp16`: True - `dataloader_num_workers`: 4 - `ddp_find_unused_parameters`: True - `batch_sampler`: no_duplicates #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: steps - `prediction_loss_only`: True - `per_device_train_batch_size`: 128 - `per_device_eval_batch_size`: 128 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 2 - `eval_accumulation_steps`: None - `torch_empty_cache_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1.0 - `num_train_epochs`: 5 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.05 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: False - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `restore_callback_states_from_checkpoint`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: True - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: True - `dataloader_num_workers`: 4 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `tp_size`: 0 - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: True - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: None - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `include_for_metrics`: [] - `eval_do_concat_batches`: True - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_eval_metrics`: False - `eval_on_start`: False - `use_liger_kernel`: False - `eval_use_gather_object`: False - `average_tokens_across_devices`: False - `prompts`: None - `batch_sampler`: no_duplicates - `multi_dataset_batch_sampler`: proportional
### Training Logs | Epoch | Step | Training Loss | full_en_cosine_ndcg@200 | full_es_cosine_ndcg@200 | full_de_cosine_ndcg@200 | full_zh_cosine_ndcg@200 | mix_es_cosine_ndcg@200 | mix_de_cosine_ndcg@200 | mix_zh_cosine_ndcg@200 | |:------:|:----:|:-------------:|:-----------------------:|:-----------------------:|:-----------------------:|:-----------------------:|:----------------------:|:----------------------:|:----------------------:| | -1 | -1 | - | 0.7322 | 0.4690 | 0.3853 | 0.2723 | 0.3209 | 0.2244 | 0.0919 | | 0.0021 | 1 | 23.8878 | - | - | - | - | - | - | - | | 0.2058 | 100 | 7.2098 | - | - | - | - | - | - | - | | 0.4115 | 200 | 4.2635 | 0.7800 | 0.5132 | 0.4268 | 0.2798 | 0.4372 | 0.2996 | 0.1447 | | 0.6173 | 300 | 4.1931 | - | - | - | - | - | - | - | | 0.8230 | 400 | 3.73 | 0.7863 | 0.5274 | 0.4451 | 0.2805 | 0.4762 | 0.3455 | 0.1648 | | 1.0309 | 500 | 3.3569 | - | - | - | - | - | - | - | | 1.2366 | 600 | 3.6464 | 0.7868 | 0.5372 | 0.4540 | 0.2813 | 0.5063 | 0.3794 | 0.1755 | | 1.4424 | 700 | 3.0772 | - | - | - | - | - | - | - | | 1.6481 | 800 | 3.114 | 0.7906 | 0.5391 | 0.4576 | 0.2832 | 0.5221 | 0.4047 | 0.1779 | | 1.8539 | 900 | 2.9246 | - | - | - | - | - | - | - | | 2.0617 | 1000 | 2.7479 | 0.7873 | 0.5423 | 0.4631 | 0.2871 | 0.5323 | 0.4143 | 0.1843 | | 2.2675 | 1100 | 3.049 | - | - | - | - | - | - | - | | 2.4733 | 1200 | 2.6137 | 0.7878 | 0.5418 | 0.4685 | 0.2870 | 0.5470 | 0.4339 | 0.1932 | | 2.6790 | 1300 | 2.8607 | - | - | - | - | - | - | - | | 2.8848 | 1400 | 2.7071 | 0.7889 | 0.5465 | 0.4714 | 0.2891 | 0.5504 | 0.4362 | 0.1944 | | 3.0926 | 1500 | 2.7012 | - | - | - | - | - | - | - | | 3.2984 | 1600 | 2.7423 | 0.7882 | 0.5471 | 0.4748 | 0.2868 | 0.5542 | 0.4454 | 0.1976 | | 3.5041 | 1700 | 2.5316 | - | - | - | - | - | - | - | | 3.7099 | 1800 | 2.6344 | 0.7900 | 0.5498 | 0.4763 | 0.2857 | 0.5639 | 0.4552 | 0.1954 | | 3.9156 | 1900 | 2.4983 | - | - | - | - | - | - | - | | 4.1235 | 2000 | 2.5423 | 0.7894 | 0.5499 | 0.4786 | 0.2870 | 0.5644 | 0.4576 | 0.1974 | ### Framework Versions - Python: 3.11.11 - Sentence Transformers: 4.1.0 - Transformers: 4.51.3 - PyTorch: 2.6.0+cu124 - Accelerate: 1.6.0 - Datasets: 3.5.0 - Tokenizers: 0.21.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### GISTEmbedLoss ```bibtex @misc{solatorio2024gistembed, title={GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuning}, author={Aivin V. Solatorio}, year={2024}, eprint={2402.16829}, archivePrefix={arXiv}, primaryClass={cs.LG} } ```