\n<img src=\"https://i.imgur.com/jC7kdl8.jpeg\" alt=\"TensorBlock\" style=\"width: 100%; min-width: 400px; display: block; margin: auto;\">\n</div>\n<div style=\"display: flex; justify-content: space-between; width: 100%;\">\n <div style=\"display: flex; flex-direction: column; align-items: flex-start;\">\n <p style=\"margin-top: 0.5em; margin-bottom: 0em;\">\n Feedback and support: TensorBlock's <a href=\"https://x.com/tensorblock_aoi\">Twitter/X</a>, <a href=\"https://t.me/TensorBlock\">Telegram Group</a> and <a href=\"https://x.com/tensorblock_aoi\">Discord server</a>\n </p>\n </div>\n</div>\n\n## Alibaba-NLP/gte-Qwen2-7B-instruct - GGUF\n\nThis repo contains GGUF format model files for [Alibaba-NLP/gte-Qwen2-7B-instruct](https://huggingface.co/Alibaba-NLP/gte-Qwen2-7B-instruct).\n\nThe files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).\n\n\n<div style=\"text-align: left; margin: 20px 0;\">\n <a href=\"https://tensorblock.co/waitlist/client\" style=\"display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;\">\n Run them on the TensorBlock client using your local machine ↗\n </a>\n</div>\n\n## Prompt template\n\n\n```\n<|im_start|>system\n{system_prompt}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant\n```\n\n## Model file specification\n\n| Filename | Quant type | File Size | Description |\n| -------- | ---------- | --------- | ----------- |\n| [gte-Qwen2-7B-instruct-Q2_K.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q2_K.gguf) | Q2_K | 2.807 GB | smallest, significant quality loss - not recommended for most purposes |\n| [gte-Qwen2-7B-instruct-Q3_K_S.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q3_K_S.gguf) | Q3_K_S | 3.251 GB | very small, high quality loss |\n| [gte-Qwen2-7B-instruct-Q3_K_M.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q3_K_M.gguf) | Q3_K_M | 3.545 GB | very small, high quality loss |\n| [gte-Qwen2-7B-instruct-Q3_K_L.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q3_K_L.gguf) | Q3_K_L | 3.806 GB | small, substantial quality loss |\n| [gte-Qwen2-7B-instruct-Q4_0.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q4_0.gguf) | Q4_0 | 4.125 GB | legacy; small, very high quality loss - prefer using Q3_K_M |\n| [gte-Qwen2-7B-instruct-Q4_K_S.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q4_K_S.gguf) | Q4_K_S | 4.150 GB | small, greater quality loss |\n| [gte-Qwen2-7B-instruct-Q4_K_M.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q4_K_M.gguf) | Q4_K_M | 4.360 GB | medium, balanced quality - recommended |\n| [gte-Qwen2-7B-instruct-Q5_0.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q5_0.gguf) | Q5_0 | 4.948 GB | legacy; medium, balanced quality - prefer using Q4_K_M |\n| [gte-Qwen2-7B-instruct-Q5_K_S.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q5_K_S.gguf) | Q5_K_S | 4.948 GB | large, low quality loss - recommended |\n| [gte-Qwen2-7B-instruct-Q5_K_M.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q5_K_M.gguf) | Q5_K_M | 5.069 GB | large, very low quality loss - recommended |\n| [gte-Qwen2-7B-instruct-Q6_K.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q6_K.gguf) | Q6_K | 5.822 GB | very large, extremely low quality loss |\n| [gte-Qwen2-7B-instruct-Q8_0.gguf](https://huggingface.co/tensorblock/gte-Qwen2-7B-instruct-GGUF/blob/main/gte-Qwen2-7B-instruct-Q8_0.gguf) | Q8_0 | 7.539 GB | very large, extremely low quality loss - not recommended |\n\n\n## Downloading instruction\n\n### Command line\n\nFirstly, install Huggingface Client\n\n```shell\npip install -U \"huggingface_hub[cli]\"\n```\n\nThen, downoad the individual model file the a local directory\n\n```shell\nhuggingface-cli download tensorblock/gte-Qwen2-7B-instruct-GGUF --include \"gte-Qwen2-7B-instruct-Q2_K.gguf\" --local-dir MY_LOCAL_DIR\n```\n\nIf you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:\n\n```shell\nhuggingface-cli download tensorblock/gte-Qwen2-7B-instruct-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'\n```\n"},"matched_task":{"kind":"list like","value":["SUMMARIZATION"],"string":"[\n \"SUMMARIZATION\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["BIOSSES","SCIFACT"],"string":"[\n \"BIOSSES\",\n \"SCIFACT\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"Non_BioNLP"}}},{"rowIdx":292,"cells":{"id":{"kind":"string","value":"ymelka/camembert-cosmetic-similarity-v2"},"author":{"kind":"string","value":"ymelka"},"task_category":{"kind":"string","value":"sentence-similarity"},"tags":{"kind":"list like","value":["sentence-transformers","safetensors","camembert","sentence-similarity","feature-extraction","generated_from_trainer","dataset_size:5000","loss:CoSENTLoss","arxiv:1908.10084","base_model:ymelka/camembert-cosmetic-finetuned","base_model:finetune:ymelka/camembert-cosmetic-finetuned","model-index","autotrain_compatible","text-embeddings-inference","endpoints_compatible","region:us"],"string":"[\n \"sentence-transformers\",\n \"safetensors\",\n \"camembert\",\n \"sentence-similarity\",\n \"feature-extraction\",\n \"generated_from_trainer\",\n \"dataset_size:5000\",\n \"loss:CoSENTLoss\",\n \"arxiv:1908.10084\",\n \"base_model:ymelka/camembert-cosmetic-finetuned\",\n \"base_model:finetune:ymelka/camembert-cosmetic-finetuned\",\n \"model-index\",\n \"autotrain_compatible\",\n \"text-embeddings-inference\",\n \"endpoints_compatible\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1718,"string":"1,718"},"last_modified":{"kind":"number","value":1718,"string":"1,718"},"downloads":{"kind":"number","value":10,"string":"10"},"likes":{"kind":"number","value":0,"string":"0"},"README":{"kind":"string","value":"---\nbase_model: ymelka/camembert-cosmetic-finetuned\ndatasets: []\nlanguage: []\nlibrary_name: sentence-transformers\nmetrics:\n- pearson_cosine\n- spearman_cosine\n- pearson_manhattan\n- spearman_manhattan\n- pearson_euclidean\n- spearman_euclidean\n- pearson_dot\n- spearman_dot\n- pearson_max\n- spearman_max\npipeline_tag: sentence-similarity\ntags:\n- sentence-transformers\n- sentence-similarity\n- feature-extraction\n- generated_from_trainer\n- dataset_size:5000\n- loss:CoSENTLoss\nwidget:\n- source_sentence: Un soin régulateur de pores hautement efficace, conçu pour réduire\n visiblement l'apparence des pores dilatés. Sa formule ciblée aide à affiner le\n grain de peau et à réguler la production de sébum, pour une peau plus lisse et\n uniforme. Idéal pour les peaux matures en quête de perfection.\n sentences:\n - La Crème Confort 1ères Rides de Coup D’Eclat est un soin hydratant apaisant qui\n procure une hydratation optimale à la peau tout en la régénérant. En 28 jours,\n la peau devient moins sensible et réactive, tandis que les premiers signes de\n l'âge sont corrigés et prévenus. Grâce à des ingrédients tels que l'huile de pépins\n de raisin, l'huile de macadamia et la vitamine E, cette crème redonne à la peau\n son éclat et sa vitalité, tout en lissant les ridules et affinant le grain de\n peau. Pour une utilisation externe sur le visage et le cou, cette formule non-comédogène\n convient à tous les types de peaux, y compris les peaux sensibles et réactives.\n Il est recommandé d'appliquer la crème matin et soir par un léger massage sur\n une peau démaquillée. Il est important de suivre les instructions d'utilisation\n et de ne pas dépasser la posologie recommandée.\n - La Source Micellaire Enchantée Rose D'Antan de Garancia est un produit de parapharmacie\n multifonctionnel qui nettoie, démaquille, hydrate, apaise et illumine la peau\n du visage, des yeux et des lèvres. Grâce à sa formule enrichie en actifs brevetés\n hydratants et apaisants, cette eau micellaire réduit de manière significative\n les sensations d'irritation et de picotement. Composée à 99,5% d'ingrédients d'origine\n naturelle, elle contient notamment de l'extrait de racine de chicorée, un prébiotique\n nourrissant pour le microbiote cutané. Pour l'utiliser, il suffit de tourner la\n pompe vers la gauche, d'imbiber un coton d'eau micellaire et de le passer sur\n le visage, les yeux et les lèvres sans rinçage. Laissez poser 5 secondes sur les\n yeux avant de démaquiller. Ce produit convient à tous les types de peau et est\n présenté dans un flacon pompe de 400 ml.\n - Le tonique hydratant Cosrx Hydrium est un produit de parapharmacie qui rend la\n peau plus fraîche et hydratée grâce à sa formule contenant de la vitamine B5 et\n de l'acide hyaluronique. Ce tonique hydratant agit comme une base essentielle\n pour la santé de la peau, en formant une barrière d'hydratation et en optimisant\n l'équilibre des peaux abîmées. Il convient à tous les types de peau, y compris\n les peaux sèches et à tendance acnéique. Les principaux ingrédients actifs incluent\n des acides hyaluroniques de type 6, du D-panthénol et de l'allantoïne pour une\n hydratation en profondeur et un effet apaisant sur les peaux sensibles. Pour une\n utilisation optimale, appliquez le tonique après le nettoyage du visage, en massant\n pour une meilleure absorption. Il peut également être utilisé comme masque en\n feuille, brume ou mélange nettoyant. Présenté en flacon de 150 ml, ce tonique\n hydratant est un allié idéal pour une peau fraîche et hydratée au quotidien.\n- source_sentence: Un soin hydratant et revitalisant qui apporte un éclat naturel\n à la peau. Enrichi en ingrédients nourrissants et anti-âge, ce soin aide à réduire\n les signes de fatigue et à améliorer la texture de la peau. Parfait pour revitaliser\n la peau mature et lui redonner toute sa jeunesse.\n sentences:\n - L'Alphanova Solide Exfoliant Visage est un produit naturel et végan qui purifie\n la peau et revitalise le teint. Composé d'huiles bio d'amande douce et de jojoba,\n de feuilles de verveine et de poudre de coques de noix, il convient aux peaux\n normales. Sans huile de palme ni sulfate, ce duo moussant doux et végétal offre\n une mousse généreuse et onctueuse au parfum frais de verveine. Avec 99,9% d'ingrédients\n d'origine naturelle, dont 73,4% issus de l'agriculture biologique, cet exfoliant\n visage Alphanova permet jusqu'à 100 utilisations. Pour l'utiliser, il suffit d'appliquer\n le nettoyant sur le visage humidifié, de masser délicatement en évitant le contour\n des yeux, puis de rincer abondamment. Présenté en deux formats de 75g, cet exfoliant\n offre une expérience de soin agréable et respectueuse de l'environnement.\n - Le Clarins Doux Nettoyant Moussant Apaisant est spécialement conçu pour les femmes\n ayant une peau très sèche ou sensible. Grâce à sa formule aux herbes des Alpes,\n ce nettoyant apaise et adoucit la peau tout en la protégeant des agressions extérieures.\n Enrichi en extraits de saponaire, de reine des près, d'aloé vera, de camomille\n bio et de beurre de karité, il nettoie en douceur, purifie, hydrate et apaise\n la peau. Sa texture mousse fine et onctueuse laisse la peau parfaitement nettoyée,\n douce et protégée. Ce nettoyant peut être utilisé matin et/ou soir en massant\n délicatement sur le visage et le cou, en évitant le contour des yeux. Il est recommandé\n de rincer abondamment après utilisation. Évitez le contour des yeux lors de l'application.\n Disponible en tube de 125 ml, ce nettoyant est idéal pour un nettoyage en douceur\n des peaux très sèches ou sensibles.\n - L'Eau Parfumée Bienfaisante Shiso de Roger&Gallet est un parfum unique aux notes\n fraîches et raffinées, mêlant le shiso, le petitgrain et la mandarine pour une\n sensation de fraîcheur naturelle. Enrichi en pivoine et en santal, ce parfum vert\n fusant apporte une énergie vibrante et permet de s'ouvrir à de nouveaux horizons.\n Idéal pour le corps, ce produit peut être utilisé en vaporisation pour accentuer\n son effet énergisant. Les principaux ingrédients actifs incluent l'extrait de\n feuille de Perilla ocymoides, connu pour ses propriétés revitalisantes. Il est\n recommandé de vaporiser un nuage de parfum devant soi et de le traverser pour\n profiter pleinement de ses bienfaits. Il est conseillé de ne pas utiliser ce produit\n sur une peau irritée ou lésée. Profitez de cette fragrance unique pour vous sentir\n revitalisé et plein d'énergie au quotidien.\n- source_sentence: Un nettoyant doux et hydratant, spécialement formulé pour éliminer\n les impuretés tout en apportant de l'éclat à la peau. Sa formule adaptée aux peaux\n matures aide à lutter contre les taches et les imperfections, tout en respectant\n la sensibilité de la peau.\n sentences:\n - Le nettoyant visage naturel solide Respire est spécialement conçu pour les peaux\n sensibles, offrant une formule douce et naturelle enrichie en huile de lin Bio,\n huile de tournesol Bio et beurre de karité Bio. Ces ingrédients apaisent, hydratent\n et protègent la peau, la laissant douce et saine. Sa formule sans ingrédients\n controversés convient parfaitement aux peaux sensibles. Facile à utiliser, il\n suffit de frotter doucement le nettoyant sur le visage humidifié, de masser légèrement\n la peau et de rincer. Vegan et non-testé sur les animaux, ce nettoyant est testé\n dermatologiquement et fabriqué en France. Il est idéal pour une peau apaisée et\n saine, et convient aux peaux sensibles. Il est recommandé de rincer immédiatement\n en cas de contact avec les yeux.\n - Le Gamarde Lait Nettoyant Douceur Peaux Délicates Bio est un nettoyant et démaquillant\n doux spécialement conçu pour les peaux délicates, sèches ou mixtes. Sa formule\n à base d'ingrédients naturels et biologiques, tels que l'eau de Gamarde les Bains,\n l'huile d'argan et l'huile de noisette, permet d'éliminer en douceur les impuretés\n et le maquillage tenace tout en respectant l'équilibre de la peau. Enrichi en\n huiles essentielles de Palmarosa et d'orange douce, ce lait nettoyant laisse la\n peau propre, douce et apaisée. Pour l'utiliser, il suffit d'appliquer une petite\n quantité sur le visage et le cou, puis de retirer avec un coton sec avant de tonifier\n la peau avec la Lotion Apaisante Douceur. Ce produit convient parfaitement pour\n un usage quotidien et ne présente aucune contre-indication particulière.\n - La serviette à cheveux Les Tendances D'Emma en couleur marron est un accessoire\n pratique et efficace pour sécher les cheveux en douceur. Fabriquée à partir de\n 90% de viscose de bambou et 10% de polyester, elle absorbe 4 fois mieux qu'une\n serviette classique. Son attache astucieuse permet de la maintenir en place sur\n la tête, évitant ainsi de traumatiser les cheveux lors du séchage. Idéale pour\n tous, y compris ceux qui ont opté pour des colorations naturelles, cette serviette\n simplifie la vie au quotidien. Facile à utiliser, il suffit de la placer sur la\n tête, de tourner et de glisser dans l'attache prévue. Lavable en machine, elle\n est pratique et écologique. Cette serviette à cheveux est conçue, fabriquée et\n imprimée en France dans une démarche éco-responsable. Un produit incontournable\n pour prendre soin de ses cheveux en toute simplicité.\n- source_sentence: Un soin anti-rides et régulateur de sébum, spécialement conçu pour\n traiter les rides et ridules tout en régulant l'excès de sébum. Sa formule hydratante\n et apaisante convient parfaitement aux peaux sensibles.\n sentences:\n - Le Phyt's Men Soin Anti-Rides est un fluide frais et non gras conçu pour atténuer\n les premiers signes de l'âge chez les hommes. Certifié Bio et d'origine naturelle,\n ce soin hydrate, raffermit et illumine la peau masculine. Sa formule contient\n des huiles végétales de sésame, noisette, chanvre, nigelle et beurre de karité,\n ainsi que des huiles essentielles de petit grain et géranium, pour leurs propriétés\n protectrices, apaisantes et tonifiantes. L'extrait de ginseng contribue à revitaliser\n la peau. Il est recommandé d'appliquer ce produit quotidiennement sur l'ensemble\n du visage. Ce soin est destiné à lutter contre les premiers signes de l'âge et\n est à usage externe uniquement. Il convient de noter que ce produit est déconseillé\n en cas d'allergie à l'un de ses composants.\n - L'Eau Thermale Spray Brumisateur Apaisant d'Avène est un soin essentiel pour les\n peaux sensibles, hypersensibles, allergiques et irritées. Grâce à sa composition\n unique en eau thermale d'Avène, ce spray apaise et sublime toutes les peaux, même\n les plus sensibles, en leur procurant une sensation d'apaisement, de confort et\n de bien-être. Les propriétés apaisantes et anti-irritantes de l'eau thermale d'Avène\n ont été démontrées par de nombreux travaux scientifiques, en faisant un véritable\n principe actif pour le traitement des affections cutanées. Il est recommandé pour\n les peaux atopiques, sébo-squameuses, couperosiques et sujettes aux photo-allergies.\n Les principaux ingrédients actifs de ce spray sont l'eau thermale Avène et le\n gaz (nitrogène), qui contribuent à apaiser la peau et à la protéger. Pour l'utiliser,\n il suffit de pulvériser une fine brume sur le visage. Ce soin a été testé par\n 100 utilisateurs qui ont tous apprécié ses bienfaits. Il est important de noter\n que ce produit est contre-indiqué en cas d'allergie à l'un de ses composants.\n - Le soin raffermissant corps et buste Copaïba Demain L'Empire 200ml est un produit\n de parapharmacie de haute qualité, formulé avec des ingrédients naturels et actifs\n pour offrir à la peau une hydratation, une protection et une fermeté optimales.\n Grâce à sa composition riche en huile de macadamia, beurre de babassu et autres\n actifs puissants, ce soin aide à améliorer l'élasticité de la peau, à réduire\n les rides, à prévenir les vergetures et à protéger contre les agressions extérieures.\n En utilisant ce produit quotidiennement, la peau retrouve sa jeunesse et sa vitalité,\n avec une texture douce et un parfum frais et vivifiant. Les principaux ingrédients\n actifs tels que la chitine, l'extrait végétal tropical et l'acide hyaluronique\n agissent en synergie pour rajeunir la peau et lui apporter une hydratation optimale.\n Il est recommandé d'appliquer ce soin sur tout le corps, en massant délicatement\n jusqu'à absorption complète. Il est conseillé de l'utiliser régulièrement pour\n des résultats visibles en seulement quelques mois. Il est important de noter que\n ce produit est destiné à un usage externe uniquement et qu'il est préférable de\n consulter un professionnel de la santé en cas de réaction allergique.\n- source_sentence: En complément du nettoyant et du soin, il est recommandé d'utiliser\n un masque purifiant et matifiant une à deux fois par semaine. Ce masque aidera\n à resserrer les pores, purifier la peau en profondeur et réguler l'excès de sébum\n pour un teint plus éclatant et uniforme.\n sentences:\n - La Crème Moussante Nettoyante Hydratante CeraVe est un produit développé en collaboration\n avec des dermatologues pour nettoyer, démaquiller et hydrater en douceur les peaux\n normales à sèches. Enrichie en céramides essentiels, acide hyaluronique et acides\n aminés, sa formule élimine efficacement les impuretés, la pollution et le maquillage\n longue tenue tout en restaurant la barrière cutanée. Grâce à la Technologie MVE,\n les actifs sont diffusés en continu pour une hydratation prolongée. Cette crème\n convient pour le visage et les yeux, est hypoallergénique et non-comédogène. Utilisez-la\n matin et soir sur une peau humide, faites mousser et rincez abondamment. Présentée\n en flacon pompe de 236 ml, elle laisse la peau douce, hydratée et propre sans\n laisser de résidus.\n - Le Fond de Teint Correcteur Fluide Avène en teinte miel est spécialement conçu\n pour corriger les imperfections cutanées modérées et unifier le teint de manière\n naturelle. Sa formule résistante à l'eau et à la sueur offre une haute tenue tout\n en protégeant la peau des rayons UV grâce à son indice de protection 20. Enrichi\n en pré-tocophéryl, il prévient le vieillissement photo-induit. Ce fond de teint\n contient un complexe pigmentaire photo-correcteur pour un teint homogène et lumineux.\n Il convient à tous les types de peaux sensibles, claires ou mates, et permet de\n camoufler efficacement les imperfections modérées. Pour une application optimale,\n il est recommandé de l'appliquer avec les doigts en unifiant sur l'ensemble du\n visage et du cou. Ce produit de parapharmacie est testé en centre de recherche\n dermatologique et utilisé à l'Atelier de Maquillage Médical de la Station thermale\n d'Avène.\n - Le Masque Purifiant Aromatique à l'Argile Darphin Skin Mat est un soin visage\n qui absorbe l'excès de sébum et purifie en profondeur l'épiderme. Grâce à sa formule,\n ce masque nettoie, clarifie et purifie la peau, la laissant plus fraîche et plus\n claire. Adapté à tous les types de peaux, il s'applique en fine couche sur le\n visage et le cou, en évitant le contour des yeux, et se laisse poser pendant 10\n à 15 minutes avant de rincer à l'eau tiède. Ce masque contient de l'argile, connue\n pour ses propriétés absorbantes et purifiantes, ainsi que des ingrédients aromatiques\n pour une expérience sensorielle agréable. Il est recommandé de l'utiliser une\n à deux fois par semaine pour des résultats optimaux. Il est conseillé de ne pas\n l'utiliser sur une peau irritée ou lésée, et de faire un test préalable sur une\n petite zone de la peau pour éviter toute réaction allergique. Profitez des bienfaits\n de ce masque pour retrouver une peau nette et éclatante.\nmodel-index:\n- name: SentenceTransformer based on ymelka/camembert-cosmetic-finetuned\n results:\n - task:\n type: semantic-similarity\n name: Semantic Similarity\n dataset:\n name: stsb fr dev\n type: stsb-fr-dev\n metrics:\n - type: pearson_cosine\n value: 0.88932660143579\n name: Pearson Cosine\n - type: spearman_cosine\n value: 0.9396146138613104\n name: Spearman Cosine\n - type: pearson_manhattan\n value: 0.881603142317095\n name: Pearson Manhattan\n - type: spearman_manhattan\n value: 0.9348905557483792\n name: Spearman Manhattan\n - type: pearson_euclidean\n value: 0.8813221547202659\n name: Pearson Euclidean\n - type: spearman_euclidean\n value: 0.936064706459536\n name: Spearman Euclidean\n - type: pearson_dot\n value: 0.7698727846138187\n name: Pearson Dot\n - type: spearman_dot\n value: 0.881900008532911\n name: Spearman Dot\n - type: pearson_max\n value: 0.88932660143579\n name: Pearson Max\n - type: spearman_max\n value: 0.9396146138613104\n name: Spearman Max\n---\n\n# SentenceTransformer based on ymelka/camembert-cosmetic-finetuned\n\nThis is a [sentence-transformers](https://www.SBERT.net) model finetuned from [ymelka/camembert-cosmetic-finetuned](https://huggingface.co/ymelka/camembert-cosmetic-finetuned). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.\n\n## Model Details\n\n### Model Description\n- **Model Type:** Sentence Transformer\n- **Base model:** [ymelka/camembert-cosmetic-finetuned](https://huggingface.co/ymelka/camembert-cosmetic-finetuned) <!-- at revision cd4cb90f9388340c5f02740130efd30336c08905 -->\n- **Maximum Sequence Length:** 512 tokens\n- **Output Dimensionality:** 768 tokens\n- **Similarity Function:** Cosine Similarity\n<!-- - **Training Dataset:** Unknown -->\n<!-- - **Language:** Unknown -->\n<!-- - **License:** Unknown -->\n\n### Model Sources\n\n- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)\n- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)\n- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)\n\n### Full Model Architecture\n\n```\nSentenceTransformer(\n (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: CamembertModel \n (1): Pooling({'word_embedding_dimension': 768, '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})\n)\n```\n\n## Usage\n\n### Direct Usage (Sentence Transformers)\n\nFirst install the Sentence Transformers library:\n\n```bash\npip install -U sentence-transformers\n```\n\nThen you can load this model and run inference.\n```python\nfrom sentence_transformers import SentenceTransformer\n\n# Download from the 🤗 Hub\nmodel = SentenceTransformer(\"ymelka/camembert-cosmetic-similarity-v2\")\n# Run inference\nsentences = [\n \"En complément du nettoyant et du soin, il est recommandé d'utiliser un masque purifiant et matifiant une à deux fois par semaine. Ce masque aidera à resserrer les pores, purifier la peau en profondeur et réguler l'excès de sébum pour un teint plus éclatant et uniforme.\",\n \"Le Masque Purifiant Aromatique à l'Argile Darphin Skin Mat est un soin visage qui absorbe l'excès de sébum et purifie en profondeur l'épiderme. Grâce à sa formule, ce masque nettoie, clarifie et purifie la peau, la laissant plus fraîche et plus claire. Adapté à tous les types de peaux, il s'applique en fine couche sur le visage et le cou, en évitant le contour des yeux, et se laisse poser pendant 10 à 15 minutes avant de rincer à l'eau tiède. Ce masque contient de l'argile, connue pour ses propriétés absorbantes et purifiantes, ainsi que des ingrédients aromatiques pour une expérience sensorielle agréable. Il est recommandé de l'utiliser une à deux fois par semaine pour des résultats optimaux. Il est conseillé de ne pas l'utiliser sur une peau irritée ou lésée, et de faire un test préalable sur une petite zone de la peau pour éviter toute réaction allergique. Profitez des bienfaits de ce masque pour retrouver une peau nette et éclatante.\",\n \"Le Fond de Teint Correcteur Fluide Avène en teinte miel est spécialement conçu pour corriger les imperfections cutanées modérées et unifier le teint de manière naturelle. Sa formule résistante à l'eau et à la sueur offre une haute tenue tout en protégeant la peau des rayons UV grâce à son indice de protection 20. Enrichi en pré-tocophéryl, il prévient le vieillissement photo-induit. Ce fond de teint contient un complexe pigmentaire photo-correcteur pour un teint homogène et lumineux. Il convient à tous les types de peaux sensibles, claires ou mates, et permet de camoufler efficacement les imperfections modérées. Pour une application optimale, il est recommandé de l'appliquer avec les doigts en unifiant sur l'ensemble du visage et du cou. Ce produit de parapharmacie est testé en centre de recherche dermatologique et utilisé à l'Atelier de Maquillage Médical de la Station thermale d'Avène.\",\n]\nembeddings = model.encode(sentences)\nprint(embeddings.shape)\n# [3, 768]\n\n# Get the similarity scores for the embeddings\nsimilarities = model.similarity(embeddings, embeddings)\nprint(similarities.shape)\n# [3, 3]\n```\n\n<!--\n### Direct Usage (Transformers)\n\n<details><summary>Click to see the direct usage in Transformers</summary>\n\n</details>\n-->\n\n<!--\n### Downstream Usage (Sentence Transformers)\n\nYou can finetune this model on your own dataset.\n\n<details><summary>Click to expand</summary>\n\n</details>\n-->\n\n<!--\n### Out-of-Scope Use\n\n*List how the model may foreseeably be misused and address what users ought not to do with the model.*\n-->\n\n## Evaluation\n\n### Metrics\n\n#### Semantic Similarity\n* Dataset: `stsb-fr-dev`\n* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)\n\n| Metric | Value |\n|:--------------------|:-----------|\n| pearson_cosine | 0.8893 |\n| **spearman_cosine** | **0.9396** |\n| pearson_manhattan | 0.8816 |\n| spearman_manhattan | 0.9349 |\n| pearson_euclidean | 0.8813 |\n| spearman_euclidean | 0.9361 |\n| pearson_dot | 0.7699 |\n| spearman_dot | 0.8819 |\n| pearson_max | 0.8893 |\n| spearman_max | 0.9396 |\n\n<!--\n## Bias, Risks and Limitations\n\n*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*\n-->\n\n<!--\n### Recommendations\n\n*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*\n-->\n\n## Training Details\n\n### Training Dataset\n\n#### Unnamed Dataset\n\n\n* Size: 5,000 training samples\n* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>\n* Approximate statistics based on the first 1000 samples:\n | | sentence1 | sentence2 | score |\n |:--------|:-----------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:---------------------------------------------------------------|\n | type | string | string | float |\n | details | <ul><li>min: 30 tokens</li><li>mean: 55.51 tokens</li><li>max: 90 tokens</li></ul> | <ul><li>min: 124 tokens</li><li>mean: 199.72 tokens</li><li>max: 503 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.47</li><li>max: 1.0</li></ul> |\n* Samples:\n | sentence1 | sentence2 | score |\n |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------|\n | <code>En complément, un sérum anti-imperfections peut être utilisé pour cibler spécifiquement les problèmes de peau tels que les imperfections et les rougeurs. Ce sérum aidera à purifier la peau et à réduire l'apparence des boutons.</code> | <code>Le sérum anti-imperfections Endro à base d'huile végétale de noisette et d'huiles essentielles bio est un concentré d'actifs naturels et antibactériens qui laisse la peau saine et les pores resserrés. Grâce à son action ciblée et hyper concentrée, il lutte efficacement contre les imperfections cutanées, réduisant ainsi les boutons et les rougeurs. Ce sérum convient aux adultes et aux adolescents à partir de 10 ans, et 73,35% des utilisateurs ont constaté une amélioration de leur peau en une semaine seulement. Les principaux ingrédients actifs tels que l'huile de noisette, l'huile essentielle de palmarosa et l'huile essentielle d'arbre à thé agissent en synergie pour purifier la peau et réguler l'excès de sébum. Pour une utilisation optimale, il est recommandé d'appliquer une petite goutte du sérum sur les zones à traiter le soir avant de dormir, en évitant le contour des yeux. Cependant, ce produit n'est pas adapté aux femmes enceintes ou allaitantes. En cas de contact avec les yeux, il est important de rincer abondamment et d'éviter toute exposition au soleil après application. Avec Endro Sérum Anti-Imperfections, retrouvez une peau nette et éclatante en toute simplicité.</code> | <code>0.9809522032737732</code> |\n | <code>Un soin régulateur et matifiant, idéal pour traiter les imperfections et les pores dilatés. Sa formule spécifique permettra de réduire l'apparence des imperfections tout en resserrant les pores pour une peau plus lisse et uniforme.</code> | <code>Le La Roche-Posay Effaclar MAT Soin Hydratant Sébo-Régulateur Visage Peaux Grasses est un soin spécialement conçu pour les peaux grasses sensibles sujettes à la brillance. Sa formule anti-brillance et anti-pores dilatés, grâce à l'association de Sebulyse, de microsphères absorbantes et de perlite, régule la production de sébum et matifie la peau immédiatement. Ce soin hydratant offre un effet matifiant et hydratant longue durée, tout en étant une excellente base de maquillage. Il convient aux adultes et aux adolescents, et est idéal pour les peaux à imperfections, à tendance acnéique et sujettes à la brillance. Pour une utilisation optimale, il est recommandé d'appliquer le produit matin et/ou soir sur l'ensemble du visage. Il est important de noter que ce produit est testé sous contrôle dermatologique, non comédogène et hypoallergénique.</code> | <code>0.9946829676628112</code> |\n | <code>Un complément de traitement anti-taches, conçu pour cibler spécifiquement les taches pigmentaires. Ce complément concentré en actifs éclaircissants aidera à atténuer les taches existantes et à prévenir l'apparition de nouvelles taches. Il est recommandé de l'utiliser en complément des autres soins pour une action ciblée et efficace.</code> | <code>Le Lierac Lumilogie Anti-Taches est un traitement ciblé pour les 3 types de taches cutanées : naissantes, visibles et incrustées. Grâce à sa formule innovante inspirée des techniques esthétiques combinées, ce produit agit sur les taches à tous les stades de leur développement. Enrichi en Hexyl R., Lys de mer et Extrait de plantain, il freine la production de mélanine, diminue les taches visibles et lutte contre l'incrustation de la mélanine en profondeur. De plus, les concentrés de vitamines E et B3 ainsi que les 7 hydroxy acides activent le renouvellement cellulaire pour éliminer la mélanine en surface. En résulte un teint unifié et plus uniforme dès la première utilisation, avec une efficacité prouvée dès 7 jours et une correction visible des taches dès 28 jours. Pour une utilisation optimale, appliquez 2 pressions du concentré jour le matin et du concentré nuit le soir sur l'ensemble du visage, en évitant le contour des yeux. Veillez à éviter le contour des yeux et à utiliser une protection solaire avec IP en cas d'exposition au soleil.</code> | <code>0.9939286708831788</code> |\n* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:\n ```json\n {\n \"scale\": 20.0,\n \"similarity_fct\": \"pairwise_cos_sim\"\n }\n ```\n\n### Evaluation Dataset\n\n#### Unnamed Dataset\n\n\n* Size: 1,000 evaluation samples\n* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>\n* Approximate statistics based on the first 1000 samples:\n | | sentence1 | sentence2 | score |\n |:--------|:-----------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:---------------------------------------------------------------|\n | type | string | string | float |\n | details | <ul><li>min: 30 tokens</li><li>mean: 54.83 tokens</li><li>max: 88 tokens</li></ul> | <ul><li>min: 120 tokens</li><li>mean: 197.93 tokens</li><li>max: 491 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.46</li><li>max: 1.0</li></ul> |\n* Samples:\n | sentence1 | sentence2 | score |\n |:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------|\n | <code>Un complément hydratant et correcteur, idéal pour les peaux sensibles et sujettes aux taches. Ce complément aidera à hydrater en profondeur, à atténuer les rides et à réduire l'apparence des pores pour une peau plus lisse et uniforme.</code> | <code>L'Huile Végétale de Karité Bio de Puressentiel est un produit nutritif, réparateur et apaisant, idéal pour nourrir et réparer en profondeur la peau et les cheveux. Cette huile 100% pure et naturelle est recommandée pour une utilisation externe sur la peau et les cheveux. Enrichie en beurre de karité issu de l'agriculture biologique, elle offre des propriétés nourrissantes et réparatrices. Pour une utilisation sur le visage et le corps, il est conseillé de chauffer une noisette de beurre de karité dans la paume de la main et de l'appliquer sur les zones sèches et craquelées. Pour les cheveux secs et abîmés, il suffit de chauffer une petite noisette de beurre de karité entre les mains et de l'appliquer sur les pointes et les longueurs. Il est important d'éviter le contact avec les yeux et les muqueuses, et de se laver les mains après application. Il est recommandé de conserver le produit à l'abri de la lumière, de l'air et de la chaleur. Disponible en pot de 100 ml, cette huile de karité bio est un allié naturel pour prendre soin de sa peau et de ses cheveux.</code> | <code>0.0544042661786079</code> |\n | <code>Un soin anti-âge global, conçu pour traiter les rides, les taches pigmentaires et les imperfections. Sa formule régulatrice et éclatante aidera à lisser la peau, à atténuer les taches et à réduire les imperfections pour un teint plus uniforme et lumineux.</code> | <code>Le sérum contour des yeux anti-rides Maison Éole Elle Et Lui Émerveillé est un produit de parapharmacie haut de gamme qui agit efficacement contre les rides, les ridules, les cernes et les poches. Sa formule complète enrichie en Bakuchiol, alternative naturelle au Rétinol A, nourrit la peau en profondeur et réduit les tâches cutanées. Grâce à ses actifs tels que l'huile de pépin de raisin, le Bisabolol et le Resvératrol, ce sérum hydrate intensément, lisse la peau et prévient le vieillissement cutané. Son utilisation matin et soir sur une peau propre permet d'obtenir un regard éclatant et reposé. Le flacon-pipette de 15ml facilite son application. Ce produit convient à tous les types de peau et ne contient ni parabène, ni silicone, ni ingrédients d'origine animale. Il est recommandé de suivre les instructions d'utilisation pour des résultats optimaux.</code> | <code>0.0781720206141471</code> |\n | <code>Un soin anti-rides et éclat, enrichi en actifs régénérants et illuminants. Ce soin aidera à lisser les rides, à uniformiser le teint et à redonner de l'éclat à la peau fatiguée.</code> | <code>L'Eau Micellaire Sebiaclear de SVR est un produit de parapharmacie qui purifie, nettoie et démaquille la peau en un seul geste. Adaptée aux peaux sensibles mixtes à grasses, cette eau micellaire aide à éliminer les impuretés, les boutons, les points noirs et l'excès de sébum sans dessécher la peau. Grâce à sa formule innovante contenant de la gluconolactone et de la niacinamide, elle offre une haute efficacité tout en respectant la peau. Les micelles présentes dans le produit nettoient et démaquillent en douceur, laissant la peau nette et fraîche. Pour l'utiliser, il suffit d'appliquer l'eau micellaire matin et/ou soir à l'aide d'un coton sur le visage et les yeux, sans rinçage. Avec une présentation en flacon de 400 ml, ce produit convient aux peaux sensibles à tendance acnéique et offre des résultats visibles dès 7 jours d'utilisation. Il est recommandé de ne pas l'utiliser en cas d'allergie à l'un des ingrédients et de consulter un professionnel de santé en cas de doute.</code> | <code>0.0607918016612529</code> |\n* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:\n ```json\n {\n \"scale\": 20.0,\n \"similarity_fct\": \"pairwise_cos_sim\"\n }\n ```\n\n### Training Hyperparameters\n#### Non-Default Hyperparameters\n\n- `eval_strategy`: steps\n- `per_device_train_batch_size`: 16\n- `per_device_eval_batch_size`: 16\n- `learning_rate`: 2e-05\n- `weight_decay`: 0.01\n- `num_train_epochs`: 4\n- `warmup_ratio`: 0.1\n- `bf16`: True\n- `load_best_model_at_end`: True\n- `batch_sampler`: no_duplicates\n\n#### All Hyperparameters\n<details><summary>Click to expand</summary>\n\n- `overwrite_output_dir`: False\n- `do_predict`: False\n- `eval_strategy`: steps\n- `prediction_loss_only`: True\n- `per_device_train_batch_size`: 16\n- `per_device_eval_batch_size`: 16\n- `per_gpu_train_batch_size`: None\n- `per_gpu_eval_batch_size`: None\n- `gradient_accumulation_steps`: 1\n- `eval_accumulation_steps`: None\n- `learning_rate`: 2e-05\n- `weight_decay`: 0.01\n- `adam_beta1`: 0.9\n- `adam_beta2`: 0.999\n- `adam_epsilon`: 1e-08\n- `max_grad_norm`: 1.0\n- `num_train_epochs`: 4\n- `max_steps`: -1\n- `lr_scheduler_type`: linear\n- `lr_scheduler_kwargs`: {}\n- `warmup_ratio`: 0.1\n- `warmup_steps`: 0\n- `log_level`: passive\n- `log_level_replica`: warning\n- `log_on_each_node`: True\n- `logging_nan_inf_filter`: True\n- `save_safetensors`: True\n- `save_on_each_node`: False\n- `save_only_model`: False\n- `restore_callback_states_from_checkpoint`: False\n- `no_cuda`: False\n- `use_cpu`: False\n- `use_mps_device`: False\n- `seed`: 42\n- `data_seed`: None\n- `jit_mode_eval`: False\n- `use_ipex`: False\n- `bf16`: True\n- `fp16`: False\n- `fp16_opt_level`: O1\n- `half_precision_backend`: auto\n- `bf16_full_eval`: False\n- `fp16_full_eval`: False\n- `tf32`: None\n- `local_rank`: 0\n- `ddp_backend`: None\n- `tpu_num_cores`: None\n- `tpu_metrics_debug`: False\n- `debug`: []\n- `dataloader_drop_last`: False\n- `dataloader_num_workers`: 0\n- `dataloader_prefetch_factor`: None\n- `past_index`: -1\n- `disable_tqdm`: False\n- `remove_unused_columns`: True\n- `label_names`: None\n- `load_best_model_at_end`: True\n- `ignore_data_skip`: False\n- `fsdp`: []\n- `fsdp_min_num_params`: 0\n- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}\n- `fsdp_transformer_layer_cls_to_wrap`: None\n- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}\n- `deepspeed`: None\n- `label_smoothing_factor`: 0.0\n- `optim`: adamw_torch\n- `optim_args`: None\n- `adafactor`: False\n- `group_by_length`: False\n- `length_column_name`: length\n- `ddp_find_unused_parameters`: None\n- `ddp_bucket_cap_mb`: None\n- `ddp_broadcast_buffers`: False\n- `dataloader_pin_memory`: True\n- `dataloader_persistent_workers`: False\n- `skip_memory_metrics`: True\n- `use_legacy_prediction_loop`: False\n- `push_to_hub`: False\n- `resume_from_checkpoint`: None\n- `hub_model_id`: None\n- `hub_strategy`: every_save\n- `hub_private_repo`: False\n- `hub_always_push`: False\n- `gradient_checkpointing`: False\n- `gradient_checkpointing_kwargs`: None\n- `include_inputs_for_metrics`: False\n- `eval_do_concat_batches`: True\n- `fp16_backend`: auto\n- `push_to_hub_model_id`: None\n- `push_to_hub_organization`: None\n- `mp_parameters`: \n- `auto_find_batch_size`: False\n- `full_determinism`: False\n- `torchdynamo`: None\n- `ray_scope`: last\n- `ddp_timeout`: 1800\n- `torch_compile`: False\n- `torch_compile_backend`: None\n- `torch_compile_mode`: None\n- `dispatch_batches`: None\n- `split_batches`: None\n- `include_tokens_per_second`: False\n- `include_num_input_tokens_seen`: False\n- `neftune_noise_alpha`: None\n- `optim_target_modules`: None\n- `batch_eval_metrics`: False\n- `batch_sampler`: no_duplicates\n- `multi_dataset_batch_sampler`: proportional\n\n</details>\n\n### Training Logs\n| Epoch | Step | Training Loss | loss | stsb-fr-dev_spearman_cosine |\n|:----------:|:-------:|:-------------:|:----------:|:---------------------------:|\n| 0 | 0 | - | - | 0.4986 |\n| 0.3195 | 100 | 4.6554 | 4.3185 | 0.8719 |\n| 0.6390 | 200 | 4.2773 | 4.1772 | 0.8984 |\n| 0.9585 | 300 | 4.1015 | 4.0808 | 0.9128 |\n| 1.2748 | 400 | 4.0285 | 4.0244 | 0.9215 |\n| 1.5942 | 500 | 3.9269 | 4.0512 | 0.9317 |\n| 1.9137 | 600 | 3.8057 | 3.9970 | 0.9348 |\n| 2.2300 | 700 | 3.7665 | 4.0250 | 0.9350 |\n| **2.5495** | **800** | **3.7541** | **3.9587** | **0.9396** |\n| 2.8690 | 900 | 3.6029 | 4.0481 | 0.9407 |\n| 3.1853 | 1000 | 3.6183 | 3.9964 | 0.9416 |\n| 3.5048 | 1100 | 3.5848 | 3.9711 | 0.9454 |\n| 3.8243 | 1200 | 3.5029 | 3.9985 | 0.9452 |\n| 3.9904 | 1252 | - | - | 0.9396 |\n\n* The bold row denotes the saved checkpoint.\n\n### Framework Versions\n- Python: 3.10.12\n- Sentence Transformers: 3.0.1\n- Transformers: 4.41.2\n- PyTorch: 2.3.0+cu121\n- Accelerate: 0.31.0\n- Datasets: 2.20.0\n- Tokenizers: 0.19.1\n\n## Citation\n\n### BibTeX\n\n#### Sentence Transformers\n```bibtex\n@inproceedings{reimers-2019-sentence-bert,\n title = \"Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks\",\n author = \"Reimers, Nils and Gurevych, Iryna\",\n booktitle = \"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing\",\n month = \"11\",\n year = \"2019\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://arxiv.org/abs/1908.10084\",\n}\n```\n\n#### CoSENTLoss\n```bibtex\n@online{kexuefm-8847,\n title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},\n author={Su Jianlin},\n year={2022},\n month={Jan},\n url={https://kexue.fm/archives/8847},\n}\n```\n\n<!--\n## Glossary\n\n*Clearly define terms in order to be accessible across audiences.*\n-->\n\n<!--\n## Model Card Authors\n\n*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*\n-->\n\n<!--\n## Model Card Contact\n\n*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*\n-->"},"matched_task":{"kind":"list like","value":["TEXT_CLASSIFICATION","SEMANTIC_SIMILARITY"],"string":"[\n \"TEXT_CLASSIFICATION\",\n \"SEMANTIC_SIMILARITY\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["CAS"],"string":"[\n \"CAS\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"Non_BioNLP"}}},{"rowIdx":293,"cells":{"id":{"kind":"string","value":"CyberOptic/multilingual-e5-large-Q8_0-GGUF"},"author":{"kind":"string","value":"CyberOptic"},"task_category":{"kind":"string","value":"feature-extraction"},"tags":{"kind":"list like","value":["sentence-transformers","gguf","mteb","Sentence Transformers","sentence-similarity","feature-extraction","llama-cpp","gguf-my-repo","multilingual","af","am","ar","as","az","be","bg","bn","br","bs","ca","cs","cy","da","de","el","en","eo","es","et","eu","fa","fi","fr","fy","ga","gd","gl","gu","ha","he","hi","hr","hu","hy","id","is","it","ja","jv","ka","kk","km","kn","ko","ku","ky","la","lo","lt","lv","mg","mk","ml","mn","mr","ms","my","ne","nl","no","om","or","pa","pl","ps","pt","ro","ru","sa","sd","si","sk","sl","so","sq","sr","su","sv","sw","ta","te","th","tl","tr","ug","uk","ur","uz","vi","xh","yi","zh","base_model:intfloat/multilingual-e5-large","base_model:quantized:intfloat/multilingual-e5-large","license:mit","model-index","autotrain_compatible","endpoints_compatible","region:us"],"string":"[\n \"sentence-transformers\",\n \"gguf\",\n \"mteb\",\n \"Sentence Transformers\",\n \"sentence-similarity\",\n \"feature-extraction\",\n \"llama-cpp\",\n \"gguf-my-repo\",\n \"multilingual\",\n \"af\",\n \"am\",\n \"ar\",\n \"as\",\n \"az\",\n \"be\",\n \"bg\",\n \"bn\",\n \"br\",\n \"bs\",\n \"ca\",\n \"cs\",\n \"cy\",\n \"da\",\n \"de\",\n \"el\",\n \"en\",\n \"eo\",\n \"es\",\n \"et\",\n \"eu\",\n \"fa\",\n \"fi\",\n \"fr\",\n \"fy\",\n \"ga\",\n \"gd\",\n \"gl\",\n \"gu\",\n \"ha\",\n \"he\",\n \"hi\",\n \"hr\",\n \"hu\",\n \"hy\",\n \"id\",\n \"is\",\n \"it\",\n \"ja\",\n \"jv\",\n \"ka\",\n \"kk\",\n \"km\",\n \"kn\",\n \"ko\",\n \"ku\",\n \"ky\",\n \"la\",\n \"lo\",\n \"lt\",\n \"lv\",\n \"mg\",\n \"mk\",\n \"ml\",\n \"mn\",\n \"mr\",\n \"ms\",\n \"my\",\n \"ne\",\n \"nl\",\n \"no\",\n \"om\",\n \"or\",\n \"pa\",\n \"pl\",\n \"ps\",\n \"pt\",\n \"ro\",\n \"ru\",\n \"sa\",\n \"sd\",\n \"si\",\n \"sk\",\n \"sl\",\n \"so\",\n \"sq\",\n \"sr\",\n \"su\",\n \"sv\",\n \"sw\",\n \"ta\",\n \"te\",\n \"th\",\n \"tl\",\n \"tr\",\n \"ug\",\n \"uk\",\n \"ur\",\n \"uz\",\n \"vi\",\n \"xh\",\n \"yi\",\n \"zh\",\n \"base_model:intfloat/multilingual-e5-large\",\n \"base_model:quantized:intfloat/multilingual-e5-large\",\n \"license:mit\",\n \"model-index\",\n \"autotrain_compatible\",\n \"endpoints_compatible\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1741,"string":"1,741"},"last_modified":{"kind":"number","value":1741,"string":"1,741"},"downloads":{"kind":"number","value":8,"string":"8"},"likes":{"kind":"number","value":0,"string":"0"},"README":{"kind":"string","value":"---\nbase_model: intfloat/multilingual-e5-large\nlanguage:\n- multilingual\n- af\n- am\n- ar\n- as\n- az\n- be\n- bg\n- bn\n- br\n- bs\n- ca\n- cs\n- cy\n- da\n- de\n- el\n- en\n- eo\n- es\n- et\n- eu\n- fa\n- fi\n- fr\n- fy\n- ga\n- gd\n- gl\n- gu\n- ha\n- he\n- hi\n- hr\n- hu\n- hy\n- id\n- is\n- it\n- ja\n- jv\n- ka\n- kk\n- km\n- kn\n- ko\n- ku\n- ky\n- la\n- lo\n- lt\n- lv\n- mg\n- mk\n- ml\n- mn\n- mr\n- ms\n- my\n- ne\n- nl\n- 'no'\n- om\n- or\n- pa\n- pl\n- ps\n- pt\n- ro\n- ru\n- sa\n- sd\n- si\n- sk\n- sl\n- so\n- sq\n- sr\n- su\n- sv\n- sw\n- ta\n- te\n- th\n- tl\n- tr\n- ug\n- uk\n- ur\n- uz\n- vi\n- xh\n- yi\n- zh\nlicense: mit\ntags:\n- mteb\n- Sentence Transformers\n- sentence-similarity\n- feature-extraction\n- sentence-transformers\n- llama-cpp\n- gguf-my-repo\nmodel-index:\n- name: multilingual-e5-large\n results:\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonCounterfactualClassification (en)\n type: mteb/amazon_counterfactual\n config: en\n split: test\n revision: e8379541af4e31359cca9fbcf4b00f2671dba205\n metrics:\n - type: accuracy\n value: 79.05970149253731\n - type: ap\n value: 43.486574390835635\n - type: f1\n value: 73.32700092140148\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonCounterfactualClassification (de)\n type: mteb/amazon_counterfactual\n config: de\n split: test\n revision: e8379541af4e31359cca9fbcf4b00f2671dba205\n metrics:\n - type: accuracy\n value: 71.22055674518201\n - type: ap\n value: 81.55756710830498\n - type: f1\n value: 69.28271787752661\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonCounterfactualClassification (en-ext)\n type: mteb/amazon_counterfactual\n config: en-ext\n split: test\n revision: e8379541af4e31359cca9fbcf4b00f2671dba205\n metrics:\n - type: accuracy\n value: 80.41979010494754\n - type: ap\n value: 29.34879922376344\n - type: f1\n value: 67.62475449011278\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonCounterfactualClassification (ja)\n type: mteb/amazon_counterfactual\n config: ja\n split: test\n revision: e8379541af4e31359cca9fbcf4b00f2671dba205\n metrics:\n - type: accuracy\n value: 77.8372591006424\n - type: ap\n value: 26.557560591210738\n - type: f1\n value: 64.96619417368707\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonPolarityClassification\n type: mteb/amazon_polarity\n config: default\n split: test\n revision: e2d317d38cd51312af73b3d32a06d1a08b442046\n metrics:\n - type: accuracy\n value: 93.489875\n - type: ap\n value: 90.98758636917603\n - type: f1\n value: 93.48554819717332\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (en)\n type: mteb/amazon_reviews_multi\n config: en\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 47.564\n - type: f1\n value: 46.75122173518047\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (de)\n type: mteb/amazon_reviews_multi\n config: de\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 45.400000000000006\n - type: f1\n value: 44.17195682400632\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (es)\n type: mteb/amazon_reviews_multi\n config: es\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 43.068\n - type: f1\n value: 42.38155696855596\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (fr)\n type: mteb/amazon_reviews_multi\n config: fr\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 41.89\n - type: f1\n value: 40.84407321682663\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (ja)\n type: mteb/amazon_reviews_multi\n config: ja\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 40.120000000000005\n - type: f1\n value: 39.522976223819114\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (zh)\n type: mteb/amazon_reviews_multi\n config: zh\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 38.832\n - type: f1\n value: 38.0392533394713\n - task:\n type: Retrieval\n dataset:\n name: MTEB ArguAna\n type: arguana\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 30.725\n - type: map_at_10\n value: 46.055\n - type: map_at_100\n value: 46.900999999999996\n - type: map_at_1000\n value: 46.911\n - type: map_at_3\n value: 41.548\n - type: map_at_5\n value: 44.297\n - type: mrr_at_1\n value: 31.152\n - type: mrr_at_10\n value: 46.231\n - type: mrr_at_100\n value: 47.07\n - type: mrr_at_1000\n value: 47.08\n - type: mrr_at_3\n value: 41.738\n - type: mrr_at_5\n value: 44.468999999999994\n - type: ndcg_at_1\n value: 30.725\n - type: ndcg_at_10\n value: 54.379999999999995\n - type: ndcg_at_100\n value: 58.138\n - type: ndcg_at_1000\n value: 58.389\n - type: ndcg_at_3\n value: 45.156\n - type: ndcg_at_5\n value: 50.123\n - type: precision_at_1\n value: 30.725\n - type: precision_at_10\n value: 8.087\n - type: precision_at_100\n value: 0.9769999999999999\n - type: precision_at_1000\n value: 0.1\n - type: precision_at_3\n value: 18.54\n - type: precision_at_5\n value: 13.542000000000002\n - type: recall_at_1\n value: 30.725\n - type: recall_at_10\n value: 80.868\n - type: recall_at_100\n value: 97.653\n - type: recall_at_1000\n value: 99.57300000000001\n - type: recall_at_3\n value: 55.619\n - type: recall_at_5\n value: 67.71000000000001\n - task:\n type: Clustering\n dataset:\n name: MTEB ArxivClusteringP2P\n type: mteb/arxiv-clustering-p2p\n config: default\n split: test\n revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d\n metrics:\n - type: v_measure\n value: 44.30960650674069\n - task:\n type: Clustering\n dataset:\n name: MTEB ArxivClusteringS2S\n type: mteb/arxiv-clustering-s2s\n config: default\n split: test\n revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53\n metrics:\n - type: v_measure\n value: 38.427074197498996\n - task:\n type: Reranking\n dataset:\n name: MTEB AskUbuntuDupQuestions\n type: mteb/askubuntudupquestions-reranking\n config: default\n split: test\n revision: 2000358ca161889fa9c082cb41daa8dcfb161a54\n metrics:\n - type: map\n value: 60.28270056031872\n - type: mrr\n value: 74.38332673789738\n - task:\n type: STS\n dataset:\n name: MTEB BIOSSES\n type: mteb/biosses-sts\n config: default\n split: test\n revision: d3fb88f8f02e40887cd149695127462bbcf29b4a\n metrics:\n - type: cos_sim_pearson\n value: 84.05942144105269\n - type: cos_sim_spearman\n value: 82.51212105850809\n - type: euclidean_pearson\n value: 81.95639829909122\n - type: euclidean_spearman\n value: 82.3717564144213\n - type: manhattan_pearson\n value: 81.79273425468256\n - type: manhattan_spearman\n value: 82.20066817871039\n - task:\n type: BitextMining\n dataset:\n name: MTEB BUCC (de-en)\n type: mteb/bucc-bitext-mining\n config: de-en\n split: test\n revision: d51519689f32196a32af33b075a01d0e7c51e252\n metrics:\n - type: accuracy\n value: 99.46764091858039\n - type: f1\n value: 99.37717466945023\n - type: precision\n value: 99.33194154488518\n - type: recall\n value: 99.46764091858039\n - task:\n type: BitextMining\n dataset:\n name: MTEB BUCC (fr-en)\n type: mteb/bucc-bitext-mining\n config: fr-en\n split: test\n revision: d51519689f32196a32af33b075a01d0e7c51e252\n metrics:\n - type: accuracy\n value: 98.29407880255337\n - type: f1\n value: 98.11248073959938\n - type: precision\n value: 98.02443319392472\n - type: recall\n value: 98.29407880255337\n - task:\n type: BitextMining\n dataset:\n name: MTEB BUCC (ru-en)\n type: mteb/bucc-bitext-mining\n config: ru-en\n split: test\n revision: d51519689f32196a32af33b075a01d0e7c51e252\n metrics:\n - type: accuracy\n value: 97.79009352268791\n - type: f1\n value: 97.5176076665512\n - type: precision\n value: 97.38136473848286\n - type: recall\n value: 97.79009352268791\n - task:\n type: BitextMining\n dataset:\n name: MTEB BUCC (zh-en)\n type: mteb/bucc-bitext-mining\n config: zh-en\n split: test\n revision: d51519689f32196a32af33b075a01d0e7c51e252\n metrics:\n - type: accuracy\n value: 99.26276987888363\n - type: f1\n value: 99.20133403545726\n - type: precision\n value: 99.17500438827453\n - type: recall\n value: 99.26276987888363\n - task:\n type: Classification\n dataset:\n name: MTEB Banking77Classification\n type: mteb/banking77\n config: default\n split: test\n revision: 0fd18e25b25c072e09e0d92ab615fda904d66300\n metrics:\n - type: accuracy\n value: 84.72727272727273\n - type: f1\n value: 84.67672206031433\n - task:\n type: Clustering\n dataset:\n name: MTEB BiorxivClusteringP2P\n type: mteb/biorxiv-clustering-p2p\n config: default\n split: test\n revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40\n metrics:\n - type: v_measure\n value: 35.34220182511161\n - task:\n type: Clustering\n dataset:\n name: MTEB BiorxivClusteringS2S\n type: mteb/biorxiv-clustering-s2s\n config: default\n split: test\n revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908\n metrics:\n - type: v_measure\n value: 33.4987096128766\n - task:\n type: Retrieval\n dataset:\n name: MTEB CQADupstackRetrieval\n type: BeIR/cqadupstack\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 25.558249999999997\n - type: map_at_10\n value: 34.44425000000001\n - type: map_at_100\n value: 35.59833333333333\n - type: map_at_1000\n value: 35.706916666666665\n - type: map_at_3\n value: 31.691749999999995\n - type: map_at_5\n value: 33.252916666666664\n - type: mrr_at_1\n value: 30.252666666666666\n - type: mrr_at_10\n value: 38.60675\n - type: mrr_at_100\n value: 39.42666666666666\n - type: mrr_at_1000\n value: 39.48408333333334\n - type: mrr_at_3\n value: 36.17441666666665\n - type: mrr_at_5\n value: 37.56275\n - type: ndcg_at_1\n value: 30.252666666666666\n - type: ndcg_at_10\n value: 39.683\n - type: ndcg_at_100\n value: 44.68541666666667\n - type: ndcg_at_1000\n value: 46.94316666666668\n - type: ndcg_at_3\n value: 34.961749999999995\n - type: ndcg_at_5\n value: 37.215666666666664\n - type: precision_at_1\n value: 30.252666666666666\n - type: precision_at_10\n value: 6.904166666666667\n - type: precision_at_100\n value: 1.0989999999999995\n - type: precision_at_1000\n value: 0.14733333333333334\n - type: precision_at_3\n value: 16.037666666666667\n - type: precision_at_5\n value: 11.413583333333333\n - type: recall_at_1\n value: 25.558249999999997\n - type: recall_at_10\n value: 51.13341666666666\n - type: recall_at_100\n value: 73.08366666666667\n - type: recall_at_1000\n value: 88.79483333333334\n - type: recall_at_3\n value: 37.989083333333326\n - type: recall_at_5\n value: 43.787833333333325\n - task:\n type: Retrieval\n dataset:\n name: MTEB ClimateFEVER\n type: climate-fever\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 10.338\n - type: map_at_10\n value: 18.360000000000003\n - type: map_at_100\n value: 19.942\n - type: map_at_1000\n value: 20.134\n - type: map_at_3\n value: 15.174000000000001\n - type: map_at_5\n value: 16.830000000000002\n - type: mrr_at_1\n value: 23.257\n - type: mrr_at_10\n value: 33.768\n - type: mrr_at_100\n value: 34.707\n - type: mrr_at_1000\n value: 34.766000000000005\n - type: mrr_at_3\n value: 30.977\n - type: mrr_at_5\n value: 32.528\n - type: ndcg_at_1\n value: 23.257\n - type: ndcg_at_10\n value: 25.733\n - type: ndcg_at_100\n value: 32.288\n - type: ndcg_at_1000\n value: 35.992000000000004\n - type: ndcg_at_3\n value: 20.866\n - type: ndcg_at_5\n value: 22.612\n - type: precision_at_1\n value: 23.257\n - type: precision_at_10\n value: 8.124\n - type: precision_at_100\n value: 1.518\n - type: precision_at_1000\n value: 0.219\n - type: precision_at_3\n value: 15.679000000000002\n - type: precision_at_5\n value: 12.117\n - type: recall_at_1\n value: 10.338\n - type: recall_at_10\n value: 31.154\n - type: recall_at_100\n value: 54.161\n - type: recall_at_1000\n value: 75.21900000000001\n - type: recall_at_3\n value: 19.427\n - type: recall_at_5\n value: 24.214\n - task:\n type: Retrieval\n dataset:\n name: MTEB DBPedia\n type: dbpedia-entity\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 8.498\n - type: map_at_10\n value: 19.103\n - type: map_at_100\n value: 27.375\n - type: map_at_1000\n value: 28.981\n - type: map_at_3\n value: 13.764999999999999\n - type: map_at_5\n value: 15.950000000000001\n - type: mrr_at_1\n value: 65.5\n - type: mrr_at_10\n value: 74.53800000000001\n - type: mrr_at_100\n value: 74.71799999999999\n - type: mrr_at_1000\n value: 74.725\n - type: mrr_at_3\n value: 72.792\n - type: mrr_at_5\n value: 73.554\n - type: ndcg_at_1\n value: 53.37499999999999\n - type: ndcg_at_10\n value: 41.286\n - type: ndcg_at_100\n value: 45.972\n - type: ndcg_at_1000\n value: 53.123\n - type: ndcg_at_3\n value: 46.172999999999995\n - type: ndcg_at_5\n value: 43.033\n - type: precision_at_1\n value: 65.5\n - type: precision_at_10\n value: 32.725\n - type: precision_at_100\n value: 10.683\n - type: precision_at_1000\n value: 1.978\n - type: precision_at_3\n value: 50\n - type: precision_at_5\n value: 41.349999999999994\n - type: recall_at_1\n value: 8.498\n - type: recall_at_10\n value: 25.070999999999998\n - type: recall_at_100\n value: 52.383\n - type: recall_at_1000\n value: 74.91499999999999\n - type: recall_at_3\n value: 15.207999999999998\n - type: recall_at_5\n value: 18.563\n - task:\n type: Classification\n dataset:\n name: MTEB EmotionClassification\n type: mteb/emotion\n config: default\n split: test\n revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37\n metrics:\n - type: accuracy\n value: 46.5\n - type: f1\n value: 41.93833713984145\n - task:\n type: Retrieval\n dataset:\n name: MTEB FEVER\n type: fever\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 67.914\n - type: map_at_10\n value: 78.10000000000001\n - type: map_at_100\n value: 78.333\n - type: map_at_1000\n value: 78.346\n - type: map_at_3\n value: 76.626\n - type: map_at_5\n value: 77.627\n - type: mrr_at_1\n value: 72.74199999999999\n - type: mrr_at_10\n value: 82.414\n - type: mrr_at_100\n value: 82.511\n - type: mrr_at_1000\n value: 82.513\n - type: mrr_at_3\n value: 81.231\n - type: mrr_at_5\n value: 82.065\n - type: ndcg_at_1\n value: 72.74199999999999\n - type: ndcg_at_10\n value: 82.806\n - type: ndcg_at_100\n value: 83.677\n - type: ndcg_at_1000\n value: 83.917\n - type: ndcg_at_3\n value: 80.305\n - type: ndcg_at_5\n value: 81.843\n - type: precision_at_1\n value: 72.74199999999999\n - type: precision_at_10\n value: 10.24\n - type: precision_at_100\n value: 1.089\n - type: precision_at_1000\n value: 0.11299999999999999\n - type: precision_at_3\n value: 31.268\n - type: precision_at_5\n value: 19.706000000000003\n - type: recall_at_1\n value: 67.914\n - type: recall_at_10\n value: 92.889\n - type: recall_at_100\n value: 96.42699999999999\n - type: recall_at_1000\n value: 97.92\n - type: recall_at_3\n value: 86.21\n - type: recall_at_5\n value: 90.036\n - task:\n type: Retrieval\n dataset:\n name: MTEB FiQA2018\n type: fiqa\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 22.166\n - type: map_at_10\n value: 35.57\n - type: map_at_100\n value: 37.405\n - type: map_at_1000\n value: 37.564\n - type: map_at_3\n value: 30.379\n - type: map_at_5\n value: 33.324\n - type: mrr_at_1\n value: 43.519000000000005\n - type: mrr_at_10\n value: 51.556000000000004\n - type: mrr_at_100\n value: 52.344\n - type: mrr_at_1000\n value: 52.373999999999995\n - type: mrr_at_3\n value: 48.868\n - type: mrr_at_5\n value: 50.319\n - type: ndcg_at_1\n value: 43.519000000000005\n - type: ndcg_at_10\n value: 43.803\n - type: ndcg_at_100\n value: 50.468999999999994\n - type: ndcg_at_1000\n value: 53.111\n - type: ndcg_at_3\n value: 38.893\n - type: ndcg_at_5\n value: 40.653\n - type: precision_at_1\n value: 43.519000000000005\n - type: precision_at_10\n value: 12.253\n - type: precision_at_100\n value: 1.931\n - type: precision_at_1000\n value: 0.242\n - type: precision_at_3\n value: 25.617\n - type: precision_at_5\n value: 19.383\n - type: recall_at_1\n value: 22.166\n - type: recall_at_10\n value: 51.6\n - type: recall_at_100\n value: 76.574\n - type: recall_at_1000\n value: 92.192\n - type: recall_at_3\n value: 34.477999999999994\n - type: recall_at_5\n value: 41.835\n - task:\n type: Retrieval\n dataset:\n name: MTEB HotpotQA\n type: hotpotqa\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 39.041\n - type: map_at_10\n value: 62.961999999999996\n - type: map_at_100\n value: 63.79899999999999\n - type: map_at_1000\n value: 63.854\n - type: map_at_3\n value: 59.399\n - type: map_at_5\n value: 61.669\n - type: mrr_at_1\n value: 78.082\n - type: mrr_at_10\n value: 84.321\n - type: mrr_at_100\n value: 84.49600000000001\n - type: mrr_at_1000\n value: 84.502\n - type: mrr_at_3\n value: 83.421\n - type: mrr_at_5\n value: 83.977\n - type: ndcg_at_1\n value: 78.082\n - type: ndcg_at_10\n value: 71.229\n - type: ndcg_at_100\n value: 74.10900000000001\n - type: ndcg_at_1000\n value: 75.169\n - type: ndcg_at_3\n value: 66.28699999999999\n - type: ndcg_at_5\n value: 69.084\n - type: precision_at_1\n value: 78.082\n - type: precision_at_10\n value: 14.993\n - type: precision_at_100\n value: 1.7239999999999998\n - type: precision_at_1000\n value: 0.186\n - type: precision_at_3\n value: 42.737\n - type: precision_at_5\n value: 27.843\n - type: recall_at_1\n value: 39.041\n - type: recall_at_10\n value: 74.96300000000001\n - type: recall_at_100\n value: 86.199\n - type: recall_at_1000\n value: 93.228\n - type: recall_at_3\n value: 64.105\n - type: recall_at_5\n value: 69.608\n - task:\n type: Classification\n dataset:\n name: MTEB ImdbClassification\n type: mteb/imdb\n config: default\n split: test\n revision: 3d86128a09e091d6018b6d26cad27f2739fc2db7\n metrics:\n - type: accuracy\n value: 90.23160000000001\n - type: ap\n value: 85.5674856808308\n - type: f1\n value: 90.18033354786317\n - task:\n type: Retrieval\n dataset:\n name: MTEB MSMARCO\n type: msmarco\n config: default\n split: dev\n revision: None\n metrics:\n - type: map_at_1\n value: 24.091\n - type: map_at_10\n value: 36.753\n - type: map_at_100\n value: 37.913000000000004\n - type: map_at_1000\n value: 37.958999999999996\n - type: map_at_3\n value: 32.818999999999996\n - type: map_at_5\n value: 35.171\n - type: mrr_at_1\n value: 24.742\n - type: mrr_at_10\n value: 37.285000000000004\n - type: mrr_at_100\n value: 38.391999999999996\n - type: mrr_at_1000\n value: 38.431\n - type: mrr_at_3\n value: 33.440999999999995\n - type: mrr_at_5\n value: 35.75\n - type: ndcg_at_1\n value: 24.742\n - type: ndcg_at_10\n value: 43.698\n - type: ndcg_at_100\n value: 49.145\n - type: ndcg_at_1000\n value: 50.23800000000001\n - type: ndcg_at_3\n value: 35.769\n - type: ndcg_at_5\n value: 39.961999999999996\n - type: precision_at_1\n value: 24.742\n - type: precision_at_10\n value: 6.7989999999999995\n - type: precision_at_100\n value: 0.95\n - type: precision_at_1000\n value: 0.104\n - type: precision_at_3\n value: 15.096000000000002\n - type: precision_at_5\n value: 11.183\n - type: recall_at_1\n value: 24.091\n - type: recall_at_10\n value: 65.068\n - type: recall_at_100\n value: 89.899\n - type: recall_at_1000\n value: 98.16\n - type: recall_at_3\n value: 43.68\n - type: recall_at_5\n value: 53.754999999999995\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (en)\n type: mteb/mtop_domain\n config: en\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 93.66621067031465\n - type: f1\n value: 93.49622853272142\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (de)\n type: mteb/mtop_domain\n config: de\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 91.94702733164272\n - type: f1\n value: 91.17043441745282\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (es)\n type: mteb/mtop_domain\n config: es\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 92.20146764509674\n - type: f1\n value: 91.98359080555608\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (fr)\n type: mteb/mtop_domain\n config: fr\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 88.99780770435328\n - type: f1\n value: 89.19746342724068\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (hi)\n type: mteb/mtop_domain\n config: hi\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 89.78486912871998\n - type: f1\n value: 89.24578823628642\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (th)\n type: mteb/mtop_domain\n config: th\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 88.74502712477394\n - type: f1\n value: 89.00297573881542\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (en)\n type: mteb/mtop_intent\n config: en\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 77.9046967624259\n - type: f1\n value: 59.36787125785957\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (de)\n type: mteb/mtop_intent\n config: de\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 74.5280360664976\n - type: f1\n value: 57.17723440888718\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (es)\n type: mteb/mtop_intent\n config: es\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 75.44029352901934\n - type: f1\n value: 54.052855531072964\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (fr)\n type: mteb/mtop_intent\n config: fr\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 70.5606013153774\n - type: f1\n value: 52.62215934386531\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (hi)\n type: mteb/mtop_intent\n config: hi\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 73.11581211903908\n - type: f1\n value: 52.341291845645465\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (th)\n type: mteb/mtop_intent\n config: th\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 74.28933092224233\n - type: f1\n value: 57.07918745504911\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (af)\n type: mteb/amazon_massive_intent\n config: af\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 62.38063214525892\n - type: f1\n value: 59.46463723443009\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (am)\n type: mteb/amazon_massive_intent\n config: am\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 56.06926698049766\n - type: f1\n value: 52.49084283283562\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ar)\n type: mteb/amazon_massive_intent\n config: ar\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 60.74983187626093\n - type: f1\n value: 56.960640620165904\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (az)\n type: mteb/amazon_massive_intent\n config: az\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 64.86550100874243\n - type: f1\n value: 62.47370548140688\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (bn)\n type: mteb/amazon_massive_intent\n config: bn\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 63.971082716879636\n - type: f1\n value: 61.03812421957381\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (cy)\n type: mteb/amazon_massive_intent\n config: cy\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 54.98318762609282\n - type: f1\n value: 51.51207916008392\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (da)\n type: mteb/amazon_massive_intent\n config: da\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.45527908540686\n - type: f1\n value: 66.16631905400318\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (de)\n type: mteb/amazon_massive_intent\n config: de\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.32750504371216\n - type: f1\n value: 66.16755288646591\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (el)\n type: mteb/amazon_massive_intent\n config: el\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.09213180901143\n - type: f1\n value: 66.95654394661507\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (en)\n type: mteb/amazon_massive_intent\n config: en\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 73.75588433086752\n - type: f1\n value: 71.79973779656923\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (es)\n type: mteb/amazon_massive_intent\n config: es\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 70.49428379287154\n - type: f1\n value: 68.37494379215734\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (fa)\n type: mteb/amazon_massive_intent\n config: fa\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.90921318090115\n - type: f1\n value: 66.79517376481645\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (fi)\n type: mteb/amazon_massive_intent\n config: fi\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 70.12104909213181\n - type: f1\n value: 67.29448842879584\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (fr)\n type: mteb/amazon_massive_intent\n config: fr\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.34095494283793\n - type: f1\n value: 67.01134288992947\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (he)\n type: mteb/amazon_massive_intent\n config: he\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 67.61264290517822\n - type: f1\n value: 64.68730512660757\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (hi)\n type: mteb/amazon_massive_intent\n config: hi\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 67.79757901815738\n - type: f1\n value: 65.24938539425598\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (hu)\n type: mteb/amazon_massive_intent\n config: hu\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.68728984532616\n - type: f1\n value: 67.0487169762553\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (hy)\n type: mteb/amazon_massive_intent\n config: hy\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 62.07464694014795\n - type: f1\n value: 59.183532276789286\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (id)\n type: mteb/amazon_massive_intent\n config: id\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 70.04707464694015\n - type: f1\n value: 67.66829629003848\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (is)\n type: mteb/amazon_massive_intent\n config: is\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 62.42434431741762\n - type: f1\n value: 59.01617226544757\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (it)\n type: mteb/amazon_massive_intent\n config: it\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 70.53127101546738\n - type: f1\n value: 68.10033760906255\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ja)\n type: mteb/amazon_massive_intent\n config: ja\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 72.50504371217215\n - type: f1\n value: 69.74931103158923\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (jv)\n type: mteb/amazon_massive_intent\n config: jv\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 57.91190316072628\n - type: f1\n value: 54.05551136648796\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ka)\n type: mteb/amazon_massive_intent\n config: ka\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 51.78211163416275\n - type: f1\n value: 49.874888544058535\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (km)\n type: mteb/amazon_massive_intent\n config: km\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 47.017484868863484\n - type: f1\n value: 44.53364263352014\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (kn)\n type: mteb/amazon_massive_intent\n config: kn\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 62.16207128446537\n - type: f1\n value: 59.01185692320829\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ko)\n type: mteb/amazon_massive_intent\n config: ko\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.42501681237391\n - type: f1\n value: 67.13169450166086\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (lv)\n type: mteb/amazon_massive_intent\n config: lv\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 67.0780094149294\n - type: f1\n value: 64.41720167850707\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ml)\n type: mteb/amazon_massive_intent\n config: ml\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 65.57162071284466\n - type: f1\n value: 62.414138683804424\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (mn)\n type: mteb/amazon_massive_intent\n config: mn\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 61.71149966375252\n - type: f1\n value: 58.594805125087234\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ms)\n type: mteb/amazon_massive_intent\n config: ms\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 66.03900470746471\n - type: f1\n value: 63.87937257883887\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (my)\n type: mteb/amazon_massive_intent\n config: my\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 60.8776059179556\n - type: f1\n value: 57.48587618059131\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (nb)\n type: mteb/amazon_massive_intent\n config: nb\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.87895090786819\n - type: f1\n value: 66.8141299430347\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (nl)\n type: mteb/amazon_massive_intent\n config: nl\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 70.45057162071285\n - type: f1\n value: 67.46444039673516\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (pl)\n type: mteb/amazon_massive_intent\n config: pl\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 71.546738399462\n - type: f1\n value: 68.63640876702655\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (pt)\n type: mteb/amazon_massive_intent\n config: pt\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 70.72965702757229\n - type: f1\n value: 68.54119560379115\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ro)\n type: mteb/amazon_massive_intent\n config: ro\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 68.35574983187625\n - type: f1\n value: 65.88844917691927\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ru)\n type: mteb/amazon_massive_intent\n config: ru\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 71.70477471418964\n - type: f1\n value: 69.19665697061978\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (sl)\n type: mteb/amazon_massive_intent\n config: sl\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 67.0880968392737\n - type: f1\n value: 64.76962317666086\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (sq)\n type: mteb/amazon_massive_intent\n config: sq\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 65.18493611297916\n - type: f1\n value: 62.49984559035371\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (sv)\n type: mteb/amazon_massive_intent\n config: sv\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 71.75857431069265\n - type: f1\n value: 69.20053687623418\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (sw)\n type: mteb/amazon_massive_intent\n config: sw\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 58.500336247478145\n - type: f1\n value: 55.2972398687929\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ta)\n type: mteb/amazon_massive_intent\n config: ta\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 62.68997982515132\n - type: f1\n value: 59.36848202755348\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (te)\n type: mteb/amazon_massive_intent\n config: te\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 63.01950235373235\n - type: f1\n value: 60.09351954625423\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (th)\n type: mteb/amazon_massive_intent\n config: th\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 68.29186281102892\n - type: f1\n value: 67.57860496703447\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (tl)\n type: mteb/amazon_massive_intent\n config: tl\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 64.77471418964357\n - type: f1\n value: 61.913983147713836\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (tr)\n type: mteb/amazon_massive_intent\n config: tr\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.87222595830532\n - type: f1\n value: 66.03679033708141\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (ur)\n type: mteb/amazon_massive_intent\n config: ur\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 64.04505716207127\n - type: f1\n value: 61.28569169817908\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (vi)\n type: mteb/amazon_massive_intent\n config: vi\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 69.38466711499663\n - type: f1\n value: 67.20532357036844\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (zh-CN)\n type: mteb/amazon_massive_intent\n config: zh-CN\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 71.12306657700067\n - type: f1\n value: 68.91251226588182\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (zh-TW)\n type: mteb/amazon_massive_intent\n config: zh-TW\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 66.20040349697378\n - type: f1\n value: 66.02657347714175\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (af)\n type: mteb/amazon_massive_scenario\n config: af\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 68.73907195696032\n - type: f1\n value: 66.98484521791418\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (am)\n type: mteb/amazon_massive_scenario\n config: am\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 60.58843308675185\n - type: f1\n value: 58.95591723092005\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ar)\n type: mteb/amazon_massive_scenario\n config: ar\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 66.22730329522528\n - type: f1\n value: 66.0894499712115\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (az)\n type: mteb/amazon_massive_scenario\n config: az\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 66.48285137861465\n - type: f1\n value: 65.21963176785157\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (bn)\n type: mteb/amazon_massive_scenario\n config: bn\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 67.74714189643578\n - type: f1\n value: 66.8212192745412\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (cy)\n type: mteb/amazon_massive_scenario\n config: cy\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 59.09213180901143\n - type: f1\n value: 56.70735546356339\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (da)\n type: mteb/amazon_massive_scenario\n config: da\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 75.05716207128448\n - type: f1\n value: 74.8413712365364\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (de)\n type: mteb/amazon_massive_scenario\n config: de\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.69737726967047\n - type: f1\n value: 74.7664341963\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (el)\n type: mteb/amazon_massive_scenario\n config: el\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.90383322125084\n - type: f1\n value: 73.59201554448323\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (en)\n type: mteb/amazon_massive_scenario\n config: en\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 77.51176866173503\n - type: f1\n value: 77.46104434577758\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (es)\n type: mteb/amazon_massive_scenario\n config: es\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.31069266980496\n - type: f1\n value: 74.61048660675635\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (fa)\n type: mteb/amazon_massive_scenario\n config: fa\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 72.95225285810356\n - type: f1\n value: 72.33160006574627\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (fi)\n type: mteb/amazon_massive_scenario\n config: fi\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.12373907195696\n - type: f1\n value: 73.20921012557481\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (fr)\n type: mteb/amazon_massive_scenario\n config: fr\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.86684599865501\n - type: f1\n value: 73.82348774610831\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (he)\n type: mteb/amazon_massive_scenario\n config: he\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 71.40215198386012\n - type: f1\n value: 71.11945183971858\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (hi)\n type: mteb/amazon_massive_scenario\n config: hi\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 72.12844653665098\n - type: f1\n value: 71.34450495911766\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (hu)\n type: mteb/amazon_massive_scenario\n config: hu\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.52252858103566\n - type: f1\n value: 73.98878711342999\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (hy)\n type: mteb/amazon_massive_scenario\n config: hy\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 64.93611297915265\n - type: f1\n value: 63.723200467653385\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (id)\n type: mteb/amazon_massive_scenario\n config: id\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.11903160726295\n - type: f1\n value: 73.82138439467096\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (is)\n type: mteb/amazon_massive_scenario\n config: is\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 67.15198386012105\n - type: f1\n value: 66.02172193802167\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (it)\n type: mteb/amazon_massive_scenario\n config: it\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.32414256893072\n - type: f1\n value: 74.30943421170574\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ja)\n type: mteb/amazon_massive_scenario\n config: ja\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 77.46805648957633\n - type: f1\n value: 77.62808409298209\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (jv)\n type: mteb/amazon_massive_scenario\n config: jv\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 63.318762609280434\n - type: f1\n value: 62.094284066075076\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ka)\n type: mteb/amazon_massive_scenario\n config: ka\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 58.34902488231338\n - type: f1\n value: 57.12893860987984\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (km)\n type: mteb/amazon_massive_scenario\n config: km\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 50.88433086751849\n - type: f1\n value: 48.2272350802058\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (kn)\n type: mteb/amazon_massive_scenario\n config: kn\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 66.4425016812374\n - type: f1\n value: 64.61463095996173\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ko)\n type: mteb/amazon_massive_scenario\n config: ko\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 75.04707464694015\n - type: f1\n value: 75.05099199098998\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (lv)\n type: mteb/amazon_massive_scenario\n config: lv\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 70.50437121721586\n - type: f1\n value: 69.83397721096314\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ml)\n type: mteb/amazon_massive_scenario\n config: ml\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 69.94283792871553\n - type: f1\n value: 68.8704663703913\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (mn)\n type: mteb/amazon_massive_scenario\n config: mn\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 64.79488903833222\n - type: f1\n value: 63.615424063345436\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ms)\n type: mteb/amazon_massive_scenario\n config: ms\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 69.88231338264963\n - type: f1\n value: 68.57892302593237\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (my)\n type: mteb/amazon_massive_scenario\n config: my\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 63.248150638870214\n - type: f1\n value: 61.06680605338809\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (nb)\n type: mteb/amazon_massive_scenario\n config: nb\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.84196368527236\n - type: f1\n value: 74.52566464968763\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (nl)\n type: mteb/amazon_massive_scenario\n config: nl\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.8285137861466\n - type: f1\n value: 74.8853197608802\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (pl)\n type: mteb/amazon_massive_scenario\n config: pl\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 74.13248150638869\n - type: f1\n value: 74.3982040999179\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (pt)\n type: mteb/amazon_massive_scenario\n config: pt\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.49024882313383\n - type: f1\n value: 73.82153848368573\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ro)\n type: mteb/amazon_massive_scenario\n config: ro\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 71.72158708809684\n - type: f1\n value: 71.85049433180541\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ru)\n type: mteb/amazon_massive_scenario\n config: ru\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 75.137861466039\n - type: f1\n value: 75.37628348188467\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (sl)\n type: mteb/amazon_massive_scenario\n config: sl\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 71.86953597848016\n - type: f1\n value: 71.87537624521661\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (sq)\n type: mteb/amazon_massive_scenario\n config: sq\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 70.27572293207801\n - type: f1\n value: 68.80017302344231\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (sv)\n type: mteb/amazon_massive_scenario\n config: sv\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 76.09952925353059\n - type: f1\n value: 76.07992707688408\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (sw)\n type: mteb/amazon_massive_scenario\n config: sw\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 63.140551445864155\n - type: f1\n value: 61.73855010331415\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ta)\n type: mteb/amazon_massive_scenario\n config: ta\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 66.27774041694687\n - type: f1\n value: 64.83664868894539\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (te)\n type: mteb/amazon_massive_scenario\n config: te\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 66.69468728984533\n - type: f1\n value: 64.76239666920868\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (th)\n type: mteb/amazon_massive_scenario\n config: th\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.44653665097512\n - type: f1\n value: 73.14646052013873\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (tl)\n type: mteb/amazon_massive_scenario\n config: tl\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 67.71351714862139\n - type: f1\n value: 66.67212180163382\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (tr)\n type: mteb/amazon_massive_scenario\n config: tr\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.9946200403497\n - type: f1\n value: 73.87348793725525\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (ur)\n type: mteb/amazon_massive_scenario\n config: ur\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 68.15400134498992\n - type: f1\n value: 67.09433241421094\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (vi)\n type: mteb/amazon_massive_scenario\n config: vi\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 73.11365164761264\n - type: f1\n value: 73.59502539433753\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (zh-CN)\n type: mteb/amazon_massive_scenario\n config: zh-CN\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 76.82582380632145\n - type: f1\n value: 76.89992945316313\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (zh-TW)\n type: mteb/amazon_massive_scenario\n config: zh-TW\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 71.81237390719569\n - type: f1\n value: 72.36499770986265\n - task:\n type: Clustering\n dataset:\n name: MTEB MedrxivClusteringP2P\n type: mteb/medrxiv-clustering-p2p\n config: default\n split: test\n revision: e7a26af6f3ae46b30dde8737f02c07b1505bcc73\n metrics:\n - type: v_measure\n value: 31.480506569594695\n - task:\n type: Clustering\n dataset:\n name: MTEB MedrxivClusteringS2S\n type: mteb/medrxiv-clustering-s2s\n config: default\n split: test\n revision: 35191c8c0dca72d8ff3efcd72aa802307d469663\n metrics:\n - type: v_measure\n value: 29.71252128004552\n - task:\n type: Reranking\n dataset:\n name: MTEB MindSmallReranking\n type: mteb/mind_small\n config: default\n split: test\n revision: 3bdac13927fdc888b903db93b2ffdbd90b295a69\n metrics:\n - type: map\n value: 31.421396787056548\n - type: mrr\n value: 32.48155274872267\n - task:\n type: Retrieval\n dataset:\n name: MTEB NFCorpus\n type: nfcorpus\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 5.595\n - type: map_at_10\n value: 12.642000000000001\n - type: map_at_100\n value: 15.726\n - type: map_at_1000\n value: 17.061999999999998\n - type: map_at_3\n value: 9.125\n - type: map_at_5\n value: 10.866000000000001\n - type: mrr_at_1\n value: 43.344\n - type: mrr_at_10\n value: 52.227999999999994\n - type: mrr_at_100\n value: 52.898999999999994\n - type: mrr_at_1000\n value: 52.944\n - type: mrr_at_3\n value: 49.845\n - type: mrr_at_5\n value: 51.115\n - type: ndcg_at_1\n value: 41.949999999999996\n - type: ndcg_at_10\n value: 33.995\n - type: ndcg_at_100\n value: 30.869999999999997\n - type: ndcg_at_1000\n value: 39.487\n - type: ndcg_at_3\n value: 38.903999999999996\n - type: ndcg_at_5\n value: 37.236999999999995\n - type: precision_at_1\n value: 43.344\n - type: precision_at_10\n value: 25.480000000000004\n - type: precision_at_100\n value: 7.672\n - type: precision_at_1000\n value: 2.028\n - type: precision_at_3\n value: 36.636\n - type: precision_at_5\n value: 32.632\n - type: recall_at_1\n value: 5.595\n - type: recall_at_10\n value: 16.466\n - type: recall_at_100\n value: 31.226\n - type: recall_at_1000\n value: 62.778999999999996\n - type: recall_at_3\n value: 9.931\n - type: recall_at_5\n value: 12.884\n - task:\n type: Retrieval\n dataset:\n name: MTEB NQ\n type: nq\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 40.414\n - type: map_at_10\n value: 56.754000000000005\n - type: map_at_100\n value: 57.457\n - type: map_at_1000\n value: 57.477999999999994\n - type: map_at_3\n value: 52.873999999999995\n - type: map_at_5\n value: 55.175\n - type: mrr_at_1\n value: 45.278\n - type: mrr_at_10\n value: 59.192\n - type: mrr_at_100\n value: 59.650000000000006\n - type: mrr_at_1000\n value: 59.665\n - type: mrr_at_3\n value: 56.141\n - type: mrr_at_5\n value: 57.998000000000005\n - type: ndcg_at_1\n value: 45.278\n - type: ndcg_at_10\n value: 64.056\n - type: ndcg_at_100\n value: 66.89\n - type: ndcg_at_1000\n value: 67.364\n - type: ndcg_at_3\n value: 56.97\n - type: ndcg_at_5\n value: 60.719\n - type: precision_at_1\n value: 45.278\n - type: precision_at_10\n value: 9.994\n - type: precision_at_100\n value: 1.165\n - type: precision_at_1000\n value: 0.121\n - type: precision_at_3\n value: 25.512\n - type: precision_at_5\n value: 17.509\n - type: recall_at_1\n value: 40.414\n - type: recall_at_10\n value: 83.596\n - type: recall_at_100\n value: 95.72\n - type: recall_at_1000\n value: 99.24\n - type: recall_at_3\n value: 65.472\n - type: recall_at_5\n value: 74.039\n - task:\n type: Retrieval\n dataset:\n name: MTEB QuoraRetrieval\n type: quora\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 70.352\n - type: map_at_10\n value: 84.369\n - type: map_at_100\n value: 85.02499999999999\n - type: map_at_1000\n value: 85.04\n - type: map_at_3\n value: 81.42399999999999\n - type: map_at_5\n value: 83.279\n - type: mrr_at_1\n value: 81.05\n - type: mrr_at_10\n value: 87.401\n - type: mrr_at_100\n value: 87.504\n - type: mrr_at_1000\n value: 87.505\n - type: mrr_at_3\n value: 86.443\n - type: mrr_at_5\n value: 87.10799999999999\n - type: ndcg_at_1\n value: 81.04\n - type: ndcg_at_10\n value: 88.181\n - type: ndcg_at_100\n value: 89.411\n - type: ndcg_at_1000\n value: 89.507\n - type: ndcg_at_3\n value: 85.28099999999999\n - type: ndcg_at_5\n value: 86.888\n - type: precision_at_1\n value: 81.04\n - type: precision_at_10\n value: 13.406\n - type: precision_at_100\n value: 1.5350000000000001\n - type: precision_at_1000\n value: 0.157\n - type: precision_at_3\n value: 37.31\n - type: precision_at_5\n value: 24.54\n - type: recall_at_1\n value: 70.352\n - type: recall_at_10\n value: 95.358\n - type: recall_at_100\n value: 99.541\n - type: recall_at_1000\n value: 99.984\n - type: recall_at_3\n value: 87.111\n - type: recall_at_5\n value: 91.643\n - task:\n type: Clustering\n dataset:\n name: MTEB RedditClustering\n type: mteb/reddit-clustering\n config: default\n split: test\n revision: 24640382cdbf8abc73003fb0fa6d111a705499eb\n metrics:\n - type: v_measure\n value: 46.54068723291946\n - task:\n type: Clustering\n dataset:\n name: MTEB RedditClusteringP2P\n type: mteb/reddit-clustering-p2p\n config: default\n split: test\n revision: 282350215ef01743dc01b456c7f5241fa8937f16\n metrics:\n - type: v_measure\n value: 63.216287629895994\n - task:\n type: Retrieval\n dataset:\n name: MTEB SCIDOCS\n type: scidocs\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 4.023000000000001\n - type: map_at_10\n value: 10.071\n - type: map_at_100\n value: 11.892\n - type: map_at_1000\n value: 12.196\n - type: map_at_3\n value: 7.234\n - type: map_at_5\n value: 8.613999999999999\n - type: mrr_at_1\n value: 19.900000000000002\n - type: mrr_at_10\n value: 30.516\n - type: mrr_at_100\n value: 31.656000000000002\n - type: mrr_at_1000\n value: 31.723000000000003\n - type: mrr_at_3\n value: 27.400000000000002\n - type: mrr_at_5\n value: 29.270000000000003\n - type: ndcg_at_1\n value: 19.900000000000002\n - type: ndcg_at_10\n value: 17.474\n - type: ndcg_at_100\n value: 25.020999999999997\n - type: ndcg_at_1000\n value: 30.728\n - type: ndcg_at_3\n value: 16.588\n - type: ndcg_at_5\n value: 14.498\n - type: precision_at_1\n value: 19.900000000000002\n - type: precision_at_10\n value: 9.139999999999999\n - type: precision_at_100\n value: 2.011\n - type: precision_at_1000\n value: 0.33899999999999997\n - type: precision_at_3\n value: 15.667\n - type: precision_at_5\n value: 12.839999999999998\n - type: recall_at_1\n value: 4.023000000000001\n - type: recall_at_10\n value: 18.497\n - type: recall_at_100\n value: 40.8\n - type: recall_at_1000\n value: 68.812\n - type: recall_at_3\n value: 9.508\n - type: recall_at_5\n value: 12.983\n - task:\n type: STS\n dataset:\n name: MTEB SICK-R\n type: mteb/sickr-sts\n config: default\n split: test\n revision: a6ea5a8cab320b040a23452cc28066d9beae2cee\n metrics:\n - type: cos_sim_pearson\n value: 83.967008785134\n - type: cos_sim_spearman\n value: 80.23142141101837\n - type: euclidean_pearson\n value: 81.20166064704539\n - type: euclidean_spearman\n value: 80.18961335654585\n - type: manhattan_pearson\n value: 81.13925443187625\n - type: manhattan_spearman\n value: 80.07948723044424\n - task:\n type: STS\n dataset:\n name: MTEB STS12\n type: mteb/sts12-sts\n config: default\n split: test\n revision: a0d554a64d88156834ff5ae9920b964011b16384\n metrics:\n - type: cos_sim_pearson\n value: 86.94262461316023\n - type: cos_sim_spearman\n value: 80.01596278563865\n - type: euclidean_pearson\n value: 83.80799622922581\n - type: euclidean_spearman\n value: 79.94984954947103\n - type: manhattan_pearson\n value: 83.68473841756281\n - type: manhattan_spearman\n value: 79.84990707951822\n - task:\n type: STS\n dataset:\n name: MTEB STS13\n type: mteb/sts13-sts\n config: default\n split: test\n revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca\n metrics:\n - type: cos_sim_pearson\n value: 80.57346443146068\n - type: cos_sim_spearman\n value: 81.54689837570866\n - type: euclidean_pearson\n value: 81.10909881516007\n - type: euclidean_spearman\n value: 81.56746243261762\n - type: manhattan_pearson\n value: 80.87076036186582\n - type: manhattan_spearman\n value: 81.33074987964402\n - task:\n type: STS\n dataset:\n name: MTEB STS14\n type: mteb/sts14-sts\n config: default\n split: test\n revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375\n metrics:\n - type: cos_sim_pearson\n value: 79.54733787179849\n - type: cos_sim_spearman\n value: 77.72202105610411\n - type: euclidean_pearson\n value: 78.9043595478849\n - type: euclidean_spearman\n value: 77.93422804309435\n - type: manhattan_pearson\n value: 78.58115121621368\n - type: manhattan_spearman\n value: 77.62508135122033\n - task:\n type: STS\n dataset:\n name: MTEB STS15\n type: mteb/sts15-sts\n config: default\n split: test\n revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3\n metrics:\n - type: cos_sim_pearson\n value: 88.59880017237558\n - type: cos_sim_spearman\n value: 89.31088630824758\n - type: euclidean_pearson\n value: 88.47069261564656\n - type: euclidean_spearman\n value: 89.33581971465233\n - type: manhattan_pearson\n value: 88.40774264100956\n - type: manhattan_spearman\n value: 89.28657485627835\n - task:\n type: STS\n dataset:\n name: MTEB STS16\n type: mteb/sts16-sts\n config: default\n split: test\n revision: 4d8694f8f0e0100860b497b999b3dbed754a0513\n metrics:\n - type: cos_sim_pearson\n value: 84.08055117917084\n - type: cos_sim_spearman\n value: 85.78491813080304\n - type: euclidean_pearson\n value: 84.99329155500392\n - type: euclidean_spearman\n value: 85.76728064677287\n - type: manhattan_pearson\n value: 84.87947428989587\n - type: manhattan_spearman\n value: 85.62429454917464\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (ko-ko)\n type: mteb/sts17-crosslingual-sts\n config: ko-ko\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 82.14190939287384\n - type: cos_sim_spearman\n value: 82.27331573306041\n - type: euclidean_pearson\n value: 81.891896953716\n - type: euclidean_spearman\n value: 82.37695542955998\n - type: manhattan_pearson\n value: 81.73123869460504\n - type: manhattan_spearman\n value: 82.19989168441421\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (ar-ar)\n type: mteb/sts17-crosslingual-sts\n config: ar-ar\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 76.84695301843362\n - type: cos_sim_spearman\n value: 77.87790986014461\n - type: euclidean_pearson\n value: 76.91981583106315\n - type: euclidean_spearman\n value: 77.88154772749589\n - type: manhattan_pearson\n value: 76.94953277451093\n - type: manhattan_spearman\n value: 77.80499230728604\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (en-ar)\n type: mteb/sts17-crosslingual-sts\n config: en-ar\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 75.44657840482016\n - type: cos_sim_spearman\n value: 75.05531095119674\n - type: euclidean_pearson\n value: 75.88161755829299\n - type: euclidean_spearman\n value: 74.73176238219332\n - type: manhattan_pearson\n value: 75.63984765635362\n - type: manhattan_spearman\n value: 74.86476440770737\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (en-de)\n type: mteb/sts17-crosslingual-sts\n config: en-de\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 85.64700140524133\n - type: cos_sim_spearman\n value: 86.16014210425672\n - type: euclidean_pearson\n value: 86.49086860843221\n - type: euclidean_spearman\n value: 86.09729326815614\n - type: manhattan_pearson\n value: 86.43406265125513\n - type: manhattan_spearman\n value: 86.17740150939994\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (en-en)\n type: mteb/sts17-crosslingual-sts\n config: en-en\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 87.91170098764921\n - type: cos_sim_spearman\n value: 88.12437004058931\n - type: euclidean_pearson\n value: 88.81828254494437\n - type: euclidean_spearman\n value: 88.14831794572122\n - type: manhattan_pearson\n value: 88.93442183448961\n - type: manhattan_spearman\n value: 88.15254630778304\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (en-tr)\n type: mteb/sts17-crosslingual-sts\n config: en-tr\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 72.91390577997292\n - type: cos_sim_spearman\n value: 71.22979457536074\n - type: euclidean_pearson\n value: 74.40314008106749\n - type: euclidean_spearman\n value: 72.54972136083246\n - type: manhattan_pearson\n value: 73.85687539530218\n - type: manhattan_spearman\n value: 72.09500771742637\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (es-en)\n type: mteb/sts17-crosslingual-sts\n config: es-en\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 80.9301067983089\n - type: cos_sim_spearman\n value: 80.74989828346473\n - type: euclidean_pearson\n value: 81.36781301814257\n - type: euclidean_spearman\n value: 80.9448819964426\n - type: manhattan_pearson\n value: 81.0351322685609\n - type: manhattan_spearman\n value: 80.70192121844177\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (es-es)\n type: mteb/sts17-crosslingual-sts\n config: es-es\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 87.13820465980005\n - type: cos_sim_spearman\n value: 86.73532498758757\n - type: euclidean_pearson\n value: 87.21329451846637\n - type: euclidean_spearman\n value: 86.57863198601002\n - type: manhattan_pearson\n value: 87.06973713818554\n - type: manhattan_spearman\n value: 86.47534918791499\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (fr-en)\n type: mteb/sts17-crosslingual-sts\n config: fr-en\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 85.48720108904415\n - type: cos_sim_spearman\n value: 85.62221757068387\n - type: euclidean_pearson\n value: 86.1010129512749\n - type: euclidean_spearman\n value: 85.86580966509942\n - type: manhattan_pearson\n value: 86.26800938808971\n - type: manhattan_spearman\n value: 85.88902721678429\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (it-en)\n type: mteb/sts17-crosslingual-sts\n config: it-en\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 83.98021347333516\n - type: cos_sim_spearman\n value: 84.53806553803501\n - type: euclidean_pearson\n value: 84.61483347248364\n - type: euclidean_spearman\n value: 85.14191408011702\n - type: manhattan_pearson\n value: 84.75297588825967\n - type: manhattan_spearman\n value: 85.33176753669242\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (nl-en)\n type: mteb/sts17-crosslingual-sts\n config: nl-en\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_pearson\n value: 84.51856644893233\n - type: cos_sim_spearman\n value: 85.27510748506413\n - type: euclidean_pearson\n value: 85.09886861540977\n - type: euclidean_spearman\n value: 85.62579245860887\n - type: manhattan_pearson\n value: 84.93017860464607\n - type: manhattan_spearman\n value: 85.5063988898453\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (en)\n type: mteb/sts22-crosslingual-sts\n config: en\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 62.581573200584195\n - type: cos_sim_spearman\n value: 63.05503590247928\n - type: euclidean_pearson\n value: 63.652564812602094\n - type: euclidean_spearman\n value: 62.64811520876156\n - type: manhattan_pearson\n value: 63.506842893061076\n - type: manhattan_spearman\n value: 62.51289573046917\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (de)\n type: mteb/sts22-crosslingual-sts\n config: de\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 48.2248801729127\n - type: cos_sim_spearman\n value: 56.5936604678561\n - type: euclidean_pearson\n value: 43.98149464089\n - type: euclidean_spearman\n value: 56.108561882423615\n - type: manhattan_pearson\n value: 43.86880305903564\n - type: manhattan_spearman\n value: 56.04671150510166\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (es)\n type: mteb/sts22-crosslingual-sts\n config: es\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 55.17564527009831\n - type: cos_sim_spearman\n value: 64.57978560979488\n - type: euclidean_pearson\n value: 58.8818330154583\n - type: euclidean_spearman\n value: 64.99214839071281\n - type: manhattan_pearson\n value: 58.72671436121381\n - type: manhattan_spearman\n value: 65.10713416616109\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (pl)\n type: mteb/sts22-crosslingual-sts\n config: pl\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 26.772131864023297\n - type: cos_sim_spearman\n value: 34.68200792408681\n - type: euclidean_pearson\n value: 16.68082419005441\n - type: euclidean_spearman\n value: 34.83099932652166\n - type: manhattan_pearson\n value: 16.52605949659529\n - type: manhattan_spearman\n value: 34.82075801399475\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (tr)\n type: mteb/sts22-crosslingual-sts\n config: tr\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 54.42415189043831\n - type: cos_sim_spearman\n value: 63.54594264576758\n - type: euclidean_pearson\n value: 57.36577498297745\n - type: euclidean_spearman\n value: 63.111466379158074\n - type: manhattan_pearson\n value: 57.584543715873885\n - type: manhattan_spearman\n value: 63.22361054139183\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (ar)\n type: mteb/sts22-crosslingual-sts\n config: ar\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 47.55216762405518\n - type: cos_sim_spearman\n value: 56.98670142896412\n - type: euclidean_pearson\n value: 50.15318757562699\n - type: euclidean_spearman\n value: 56.524941926541906\n - type: manhattan_pearson\n value: 49.955618528674904\n - type: manhattan_spearman\n value: 56.37102209240117\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (ru)\n type: mteb/sts22-crosslingual-sts\n config: ru\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 49.20540980338571\n - type: cos_sim_spearman\n value: 59.9009453504406\n - type: euclidean_pearson\n value: 49.557749853620535\n - type: euclidean_spearman\n value: 59.76631621172456\n - type: manhattan_pearson\n value: 49.62340591181147\n - type: manhattan_spearman\n value: 59.94224880322436\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (zh)\n type: mteb/sts22-crosslingual-sts\n config: zh\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 51.508169956576985\n - type: cos_sim_spearman\n value: 66.82461565306046\n - type: euclidean_pearson\n value: 56.2274426480083\n - type: euclidean_spearman\n value: 66.6775323848333\n - type: manhattan_pearson\n value: 55.98277796300661\n - type: manhattan_spearman\n value: 66.63669848497175\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (fr)\n type: mteb/sts22-crosslingual-sts\n config: fr\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 72.86478788045507\n - type: cos_sim_spearman\n value: 76.7946552053193\n - type: euclidean_pearson\n value: 75.01598530490269\n - type: euclidean_spearman\n value: 76.83618917858281\n - type: manhattan_pearson\n value: 74.68337628304332\n - type: manhattan_spearman\n value: 76.57480204017773\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (de-en)\n type: mteb/sts22-crosslingual-sts\n config: de-en\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 55.922619099401984\n - type: cos_sim_spearman\n value: 56.599362477240774\n - type: euclidean_pearson\n value: 56.68307052369783\n - type: euclidean_spearman\n value: 54.28760436777401\n - type: manhattan_pearson\n value: 56.67763566500681\n - type: manhattan_spearman\n value: 53.94619541711359\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (es-en)\n type: mteb/sts22-crosslingual-sts\n config: es-en\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 66.74357206710913\n - type: cos_sim_spearman\n value: 72.5208244925311\n - type: euclidean_pearson\n value: 67.49254562186032\n - type: euclidean_spearman\n value: 72.02469076238683\n - type: manhattan_pearson\n value: 67.45251772238085\n - type: manhattan_spearman\n value: 72.05538819984538\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (it)\n type: mteb/sts22-crosslingual-sts\n config: it\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 71.25734330033191\n - type: cos_sim_spearman\n value: 76.98349083946823\n - type: euclidean_pearson\n value: 73.71642838667736\n - type: euclidean_spearman\n value: 77.01715504651384\n - type: manhattan_pearson\n value: 73.61712711868105\n - type: manhattan_spearman\n value: 77.01392571153896\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (pl-en)\n type: mteb/sts22-crosslingual-sts\n config: pl-en\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 63.18215462781212\n - type: cos_sim_spearman\n value: 65.54373266117607\n - type: euclidean_pearson\n value: 64.54126095439005\n - type: euclidean_spearman\n value: 65.30410369102711\n - type: manhattan_pearson\n value: 63.50332221148234\n - type: manhattan_spearman\n value: 64.3455878104313\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (zh-en)\n type: mteb/sts22-crosslingual-sts\n config: zh-en\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 62.30509221440029\n - type: cos_sim_spearman\n value: 65.99582704642478\n - type: euclidean_pearson\n value: 63.43818859884195\n - type: euclidean_spearman\n value: 66.83172582815764\n - type: manhattan_pearson\n value: 63.055779168508764\n - type: manhattan_spearman\n value: 65.49585020501449\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (es-it)\n type: mteb/sts22-crosslingual-sts\n config: es-it\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 59.587830825340404\n - type: cos_sim_spearman\n value: 68.93467614588089\n - type: euclidean_pearson\n value: 62.3073527367404\n - type: euclidean_spearman\n value: 69.69758171553175\n - type: manhattan_pearson\n value: 61.9074580815789\n - type: manhattan_spearman\n value: 69.57696375597865\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (de-fr)\n type: mteb/sts22-crosslingual-sts\n config: de-fr\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 57.143220125577066\n - type: cos_sim_spearman\n value: 67.78857859159226\n - type: euclidean_pearson\n value: 55.58225107923733\n - type: euclidean_spearman\n value: 67.80662907184563\n - type: manhattan_pearson\n value: 56.24953502726514\n - type: manhattan_spearman\n value: 67.98262125431616\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (de-pl)\n type: mteb/sts22-crosslingual-sts\n config: de-pl\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 21.826928900322066\n - type: cos_sim_spearman\n value: 49.578506634400405\n - type: euclidean_pearson\n value: 27.939890138843214\n - type: euclidean_spearman\n value: 52.71950519136242\n - type: manhattan_pearson\n value: 26.39878683847546\n - type: manhattan_spearman\n value: 47.54609580342499\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (fr-pl)\n type: mteb/sts22-crosslingual-sts\n config: fr-pl\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_pearson\n value: 57.27603854632001\n - type: cos_sim_spearman\n value: 50.709255283710995\n - type: euclidean_pearson\n value: 59.5419024445929\n - type: euclidean_spearman\n value: 50.709255283710995\n - type: manhattan_pearson\n value: 59.03256832438492\n - type: manhattan_spearman\n value: 61.97797868009122\n - task:\n type: STS\n dataset:\n name: MTEB STSBenchmark\n type: mteb/stsbenchmark-sts\n config: default\n split: test\n revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831\n metrics:\n - type: cos_sim_pearson\n value: 85.00757054859712\n - type: cos_sim_spearman\n value: 87.29283629622222\n - type: euclidean_pearson\n value: 86.54824171775536\n - type: euclidean_spearman\n value: 87.24364730491402\n - type: manhattan_pearson\n value: 86.5062156915074\n - type: manhattan_spearman\n value: 87.15052170378574\n - task:\n type: Reranking\n dataset:\n name: MTEB SciDocsRR\n type: mteb/scidocs-reranking\n config: default\n split: test\n revision: d3c5e1fc0b855ab6097bf1cda04dd73947d7caab\n metrics:\n - type: map\n value: 82.03549357197389\n - type: mrr\n value: 95.05437645143527\n - task:\n type: Retrieval\n dataset:\n name: MTEB SciFact\n type: scifact\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 57.260999999999996\n - type: map_at_10\n value: 66.259\n - type: map_at_100\n value: 66.884\n - type: map_at_1000\n value: 66.912\n - type: map_at_3\n value: 63.685\n - type: map_at_5\n value: 65.35499999999999\n - type: mrr_at_1\n value: 60.333000000000006\n - type: mrr_at_10\n value: 67.5\n - type: mrr_at_100\n value: 68.013\n - type: mrr_at_1000\n value: 68.038\n - type: mrr_at_3\n value: 65.61099999999999\n - type: mrr_at_5\n value: 66.861\n - type: ndcg_at_1\n value: 60.333000000000006\n - type: ndcg_at_10\n value: 70.41\n - type: ndcg_at_100\n value: 73.10600000000001\n - type: ndcg_at_1000\n value: 73.846\n - type: ndcg_at_3\n value: 66.133\n - type: ndcg_at_5\n value: 68.499\n - type: precision_at_1\n value: 60.333000000000006\n - type: precision_at_10\n value: 9.232999999999999\n - type: precision_at_100\n value: 1.0630000000000002\n - type: precision_at_1000\n value: 0.11299999999999999\n - type: precision_at_3\n value: 25.667\n - type: precision_at_5\n value: 17.067\n - type: recall_at_1\n value: 57.260999999999996\n - type: recall_at_10\n value: 81.94399999999999\n - type: recall_at_100\n value: 93.867\n - type: recall_at_1000\n value: 99.667\n - type: recall_at_3\n value: 70.339\n - type: recall_at_5\n value: 76.25\n - task:\n type: PairClassification\n dataset:\n name: MTEB SprintDuplicateQuestions\n type: mteb/sprintduplicatequestions-pairclassification\n config: default\n split: test\n revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46\n metrics:\n - type: cos_sim_accuracy\n value: 99.74356435643564\n - type: cos_sim_ap\n value: 93.13411948212683\n - type: cos_sim_f1\n value: 86.80521991300147\n - type: cos_sim_precision\n value: 84.00374181478017\n - type: cos_sim_recall\n value: 89.8\n - type: dot_accuracy\n value: 99.67920792079208\n - type: dot_ap\n value: 89.27277565444479\n - type: dot_f1\n value: 83.9276990718124\n - type: dot_precision\n value: 82.04393505253104\n - type: dot_recall\n value: 85.9\n - type: euclidean_accuracy\n value: 99.74257425742574\n - type: euclidean_ap\n value: 93.17993008259062\n - type: euclidean_f1\n value: 86.69396110542476\n - type: euclidean_precision\n value: 88.78406708595388\n - type: euclidean_recall\n value: 84.7\n - type: manhattan_accuracy\n value: 99.74257425742574\n - type: manhattan_ap\n value: 93.14413755550099\n - type: manhattan_f1\n value: 86.82483594144371\n - type: manhattan_precision\n value: 87.66564729867483\n - type: manhattan_recall\n value: 86\n - type: max_accuracy\n value: 99.74356435643564\n - type: max_ap\n value: 93.17993008259062\n - type: max_f1\n value: 86.82483594144371\n - task:\n type: Clustering\n dataset:\n name: MTEB StackExchangeClustering\n type: mteb/stackexchange-clustering\n config: default\n split: test\n revision: 6cbc1f7b2bc0622f2e39d2c77fa502909748c259\n metrics:\n - type: v_measure\n value: 57.525863806168566\n - task:\n type: Clustering\n dataset:\n name: MTEB StackExchangeClusteringP2P\n type: mteb/stackexchange-clustering-p2p\n config: default\n split: test\n revision: 815ca46b2622cec33ccafc3735d572c266efdb44\n metrics:\n - type: v_measure\n value: 32.68850574423839\n - task:\n type: Reranking\n dataset:\n name: MTEB StackOverflowDupQuestions\n type: mteb/stackoverflowdupquestions-reranking\n config: default\n split: test\n revision: e185fbe320c72810689fc5848eb6114e1ef5ec69\n metrics:\n - type: map\n value: 49.71580650644033\n - type: mrr\n value: 50.50971903913081\n - task:\n type: Summarization\n dataset:\n name: MTEB SummEval\n type: mteb/summeval\n config: default\n split: test\n revision: cda12ad7615edc362dbf25a00fdd61d3b1eaf93c\n metrics:\n - type: cos_sim_pearson\n value: 29.152190498799484\n - type: cos_sim_spearman\n value: 29.686180371952727\n - type: dot_pearson\n value: 27.248664793816342\n - type: dot_spearman\n value: 28.37748983721745\n - task:\n type: Retrieval\n dataset:\n name: MTEB TRECCOVID\n type: trec-covid\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 0.20400000000000001\n - type: map_at_10\n value: 1.6209999999999998\n - type: map_at_100\n value: 9.690999999999999\n - type: map_at_1000\n value: 23.733\n - type: map_at_3\n value: 0.575\n - type: map_at_5\n value: 0.885\n - type: mrr_at_1\n value: 78\n - type: mrr_at_10\n value: 86.56700000000001\n - type: mrr_at_100\n value: 86.56700000000001\n - type: mrr_at_1000\n value: 86.56700000000001\n - type: mrr_at_3\n value: 85.667\n - type: mrr_at_5\n value: 86.56700000000001\n - type: ndcg_at_1\n value: 76\n - type: ndcg_at_10\n value: 71.326\n - type: ndcg_at_100\n value: 54.208999999999996\n - type: ndcg_at_1000\n value: 49.252\n - type: ndcg_at_3\n value: 74.235\n - type: ndcg_at_5\n value: 73.833\n - type: precision_at_1\n value: 78\n - type: precision_at_10\n value: 74.8\n - type: precision_at_100\n value: 55.50000000000001\n - type: precision_at_1000\n value: 21.836\n - type: precision_at_3\n value: 78\n - type: precision_at_5\n value: 78\n - type: recall_at_1\n value: 0.20400000000000001\n - type: recall_at_10\n value: 1.894\n - type: recall_at_100\n value: 13.245999999999999\n - type: recall_at_1000\n value: 46.373\n - type: recall_at_3\n value: 0.613\n - type: recall_at_5\n value: 0.991\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (sqi-eng)\n type: mteb/tatoeba-bitext-mining\n config: sqi-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 95.89999999999999\n - type: f1\n value: 94.69999999999999\n - type: precision\n value: 94.11666666666667\n - type: recall\n value: 95.89999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (fry-eng)\n type: mteb/tatoeba-bitext-mining\n config: fry-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 68.20809248554913\n - type: f1\n value: 63.431048720066066\n - type: precision\n value: 61.69143958161298\n - type: recall\n value: 68.20809248554913\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (kur-eng)\n type: mteb/tatoeba-bitext-mining\n config: kur-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 71.21951219512195\n - type: f1\n value: 66.82926829268293\n - type: precision\n value: 65.1260162601626\n - type: recall\n value: 71.21951219512195\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tur-eng)\n type: mteb/tatoeba-bitext-mining\n config: tur-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97.2\n - type: f1\n value: 96.26666666666667\n - type: precision\n value: 95.8\n - type: recall\n value: 97.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (deu-eng)\n type: mteb/tatoeba-bitext-mining\n config: deu-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 99.3\n - type: f1\n value: 99.06666666666666\n - type: precision\n value: 98.95\n - type: recall\n value: 99.3\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (nld-eng)\n type: mteb/tatoeba-bitext-mining\n config: nld-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97.39999999999999\n - type: f1\n value: 96.63333333333333\n - type: precision\n value: 96.26666666666668\n - type: recall\n value: 97.39999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ron-eng)\n type: mteb/tatoeba-bitext-mining\n config: ron-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96\n - type: f1\n value: 94.86666666666666\n - type: precision\n value: 94.31666666666668\n - type: recall\n value: 96\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ang-eng)\n type: mteb/tatoeba-bitext-mining\n config: ang-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 47.01492537313433\n - type: f1\n value: 40.178867566927266\n - type: precision\n value: 38.179295828549556\n - type: recall\n value: 47.01492537313433\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ido-eng)\n type: mteb/tatoeba-bitext-mining\n config: ido-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 86.5\n - type: f1\n value: 83.62537480063796\n - type: precision\n value: 82.44555555555554\n - type: recall\n value: 86.5\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (jav-eng)\n type: mteb/tatoeba-bitext-mining\n config: jav-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 80.48780487804879\n - type: f1\n value: 75.45644599303138\n - type: precision\n value: 73.37398373983739\n - type: recall\n value: 80.48780487804879\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (isl-eng)\n type: mteb/tatoeba-bitext-mining\n config: isl-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 93.7\n - type: f1\n value: 91.95666666666666\n - type: precision\n value: 91.125\n - type: recall\n value: 93.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (slv-eng)\n type: mteb/tatoeba-bitext-mining\n config: slv-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 91.73754556500607\n - type: f1\n value: 89.65168084244632\n - type: precision\n value: 88.73025516403402\n - type: recall\n value: 91.73754556500607\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (cym-eng)\n type: mteb/tatoeba-bitext-mining\n config: cym-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 81.04347826086956\n - type: f1\n value: 76.2128364389234\n - type: precision\n value: 74.2\n - type: recall\n value: 81.04347826086956\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (kaz-eng)\n type: mteb/tatoeba-bitext-mining\n config: kaz-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 83.65217391304348\n - type: f1\n value: 79.4376811594203\n - type: precision\n value: 77.65797101449274\n - type: recall\n value: 83.65217391304348\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (est-eng)\n type: mteb/tatoeba-bitext-mining\n config: est-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 87.5\n - type: f1\n value: 85.02690476190476\n - type: precision\n value: 83.96261904761904\n - type: recall\n value: 87.5\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (heb-eng)\n type: mteb/tatoeba-bitext-mining\n config: heb-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 89.3\n - type: f1\n value: 86.52333333333333\n - type: precision\n value: 85.22833333333332\n - type: recall\n value: 89.3\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (gla-eng)\n type: mteb/tatoeba-bitext-mining\n config: gla-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 65.01809408926418\n - type: f1\n value: 59.00594446432805\n - type: precision\n value: 56.827215807915444\n - type: recall\n value: 65.01809408926418\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (mar-eng)\n type: mteb/tatoeba-bitext-mining\n config: mar-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 91.2\n - type: f1\n value: 88.58\n - type: precision\n value: 87.33333333333334\n - type: recall\n value: 91.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (lat-eng)\n type: mteb/tatoeba-bitext-mining\n config: lat-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 59.199999999999996\n - type: f1\n value: 53.299166276284915\n - type: precision\n value: 51.3383908045977\n - type: recall\n value: 59.199999999999996\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (bel-eng)\n type: mteb/tatoeba-bitext-mining\n config: bel-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 93.2\n - type: f1\n value: 91.2\n - type: precision\n value: 90.25\n - type: recall\n value: 93.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (pms-eng)\n type: mteb/tatoeba-bitext-mining\n config: pms-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 64.76190476190476\n - type: f1\n value: 59.867110667110666\n - type: precision\n value: 58.07390192653351\n - type: recall\n value: 64.76190476190476\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (gle-eng)\n type: mteb/tatoeba-bitext-mining\n config: gle-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 76.2\n - type: f1\n value: 71.48147546897547\n - type: precision\n value: 69.65409090909091\n - type: recall\n value: 76.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (pes-eng)\n type: mteb/tatoeba-bitext-mining\n config: pes-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 93.8\n - type: f1\n value: 92.14\n - type: precision\n value: 91.35833333333333\n - type: recall\n value: 93.8\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (nob-eng)\n type: mteb/tatoeba-bitext-mining\n config: nob-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97.89999999999999\n - type: f1\n value: 97.2\n - type: precision\n value: 96.85000000000001\n - type: recall\n value: 97.89999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (bul-eng)\n type: mteb/tatoeba-bitext-mining\n config: bul-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.6\n - type: f1\n value: 92.93333333333334\n - type: precision\n value: 92.13333333333333\n - type: recall\n value: 94.6\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (cbk-eng)\n type: mteb/tatoeba-bitext-mining\n config: cbk-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 74.1\n - type: f1\n value: 69.14817460317461\n - type: precision\n value: 67.2515873015873\n - type: recall\n value: 74.1\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (hun-eng)\n type: mteb/tatoeba-bitext-mining\n config: hun-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 95.19999999999999\n - type: f1\n value: 94.01333333333335\n - type: precision\n value: 93.46666666666667\n - type: recall\n value: 95.19999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (uig-eng)\n type: mteb/tatoeba-bitext-mining\n config: uig-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 76.9\n - type: f1\n value: 72.07523809523809\n - type: precision\n value: 70.19777777777779\n - type: recall\n value: 76.9\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (rus-eng)\n type: mteb/tatoeba-bitext-mining\n config: rus-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.1\n - type: f1\n value: 92.31666666666666\n - type: precision\n value: 91.43333333333332\n - type: recall\n value: 94.1\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (spa-eng)\n type: mteb/tatoeba-bitext-mining\n config: spa-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97.8\n - type: f1\n value: 97.1\n - type: precision\n value: 96.76666666666668\n - type: recall\n value: 97.8\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (hye-eng)\n type: mteb/tatoeba-bitext-mining\n config: hye-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 92.85714285714286\n - type: f1\n value: 90.92093441150045\n - type: precision\n value: 90.00449236298293\n - type: recall\n value: 92.85714285714286\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tel-eng)\n type: mteb/tatoeba-bitext-mining\n config: tel-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 93.16239316239316\n - type: f1\n value: 91.33903133903132\n - type: precision\n value: 90.56267806267806\n - type: recall\n value: 93.16239316239316\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (afr-eng)\n type: mteb/tatoeba-bitext-mining\n config: afr-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 92.4\n - type: f1\n value: 90.25666666666666\n - type: precision\n value: 89.25833333333334\n - type: recall\n value: 92.4\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (mon-eng)\n type: mteb/tatoeba-bitext-mining\n config: mon-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 90.22727272727272\n - type: f1\n value: 87.53030303030303\n - type: precision\n value: 86.37121212121211\n - type: recall\n value: 90.22727272727272\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (arz-eng)\n type: mteb/tatoeba-bitext-mining\n config: arz-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 79.03563941299791\n - type: f1\n value: 74.7349505840072\n - type: precision\n value: 72.9035639412998\n - type: recall\n value: 79.03563941299791\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (hrv-eng)\n type: mteb/tatoeba-bitext-mining\n config: hrv-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97\n - type: f1\n value: 96.15\n - type: precision\n value: 95.76666666666668\n - type: recall\n value: 97\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (nov-eng)\n type: mteb/tatoeba-bitext-mining\n config: nov-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 76.26459143968872\n - type: f1\n value: 71.55642023346303\n - type: precision\n value: 69.7544932369835\n - type: recall\n value: 76.26459143968872\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (gsw-eng)\n type: mteb/tatoeba-bitext-mining\n config: gsw-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 58.119658119658126\n - type: f1\n value: 51.65242165242165\n - type: precision\n value: 49.41768108434775\n - type: recall\n value: 58.119658119658126\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (nds-eng)\n type: mteb/tatoeba-bitext-mining\n config: nds-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 74.3\n - type: f1\n value: 69.52055555555555\n - type: precision\n value: 67.7574938949939\n - type: recall\n value: 74.3\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ukr-eng)\n type: mteb/tatoeba-bitext-mining\n config: ukr-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.8\n - type: f1\n value: 93.31666666666666\n - type: precision\n value: 92.60000000000001\n - type: recall\n value: 94.8\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (uzb-eng)\n type: mteb/tatoeba-bitext-mining\n config: uzb-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 76.63551401869158\n - type: f1\n value: 72.35202492211837\n - type: precision\n value: 70.60358255451713\n - type: recall\n value: 76.63551401869158\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (lit-eng)\n type: mteb/tatoeba-bitext-mining\n config: lit-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 90.4\n - type: f1\n value: 88.4811111111111\n - type: precision\n value: 87.7452380952381\n - type: recall\n value: 90.4\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ina-eng)\n type: mteb/tatoeba-bitext-mining\n config: ina-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 95\n - type: f1\n value: 93.60666666666667\n - type: precision\n value: 92.975\n - type: recall\n value: 95\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (lfn-eng)\n type: mteb/tatoeba-bitext-mining\n config: lfn-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 67.2\n - type: f1\n value: 63.01595782872099\n - type: precision\n value: 61.596587301587306\n - type: recall\n value: 67.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (zsm-eng)\n type: mteb/tatoeba-bitext-mining\n config: zsm-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 95.7\n - type: f1\n value: 94.52999999999999\n - type: precision\n value: 94\n - type: recall\n value: 95.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ita-eng)\n type: mteb/tatoeba-bitext-mining\n config: ita-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.6\n - type: f1\n value: 93.28999999999999\n - type: precision\n value: 92.675\n - type: recall\n value: 94.6\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (cmn-eng)\n type: mteb/tatoeba-bitext-mining\n config: cmn-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.39999999999999\n - type: f1\n value: 95.28333333333333\n - type: precision\n value: 94.75\n - type: recall\n value: 96.39999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (lvs-eng)\n type: mteb/tatoeba-bitext-mining\n config: lvs-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 91.9\n - type: f1\n value: 89.83\n - type: precision\n value: 88.92\n - type: recall\n value: 91.9\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (glg-eng)\n type: mteb/tatoeba-bitext-mining\n config: glg-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.69999999999999\n - type: f1\n value: 93.34222222222223\n - type: precision\n value: 92.75416666666668\n - type: recall\n value: 94.69999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ceb-eng)\n type: mteb/tatoeba-bitext-mining\n config: ceb-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 60.333333333333336\n - type: f1\n value: 55.31203703703703\n - type: precision\n value: 53.39971108326371\n - type: recall\n value: 60.333333333333336\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (bre-eng)\n type: mteb/tatoeba-bitext-mining\n config: bre-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 12.9\n - type: f1\n value: 11.099861903031458\n - type: precision\n value: 10.589187932631877\n - type: recall\n value: 12.9\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ben-eng)\n type: mteb/tatoeba-bitext-mining\n config: ben-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 86.7\n - type: f1\n value: 83.0152380952381\n - type: precision\n value: 81.37833333333333\n - type: recall\n value: 86.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (swg-eng)\n type: mteb/tatoeba-bitext-mining\n config: swg-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 63.39285714285714\n - type: f1\n value: 56.832482993197274\n - type: precision\n value: 54.56845238095237\n - type: recall\n value: 63.39285714285714\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (arq-eng)\n type: mteb/tatoeba-bitext-mining\n config: arq-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 48.73765093304062\n - type: f1\n value: 41.555736920720456\n - type: precision\n value: 39.06874531737319\n - type: recall\n value: 48.73765093304062\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (kab-eng)\n type: mteb/tatoeba-bitext-mining\n config: kab-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 41.099999999999994\n - type: f1\n value: 36.540165945165946\n - type: precision\n value: 35.05175685425686\n - type: recall\n value: 41.099999999999994\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (fra-eng)\n type: mteb/tatoeba-bitext-mining\n config: fra-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.89999999999999\n - type: f1\n value: 93.42333333333333\n - type: precision\n value: 92.75833333333333\n - type: recall\n value: 94.89999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (por-eng)\n type: mteb/tatoeba-bitext-mining\n config: por-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.89999999999999\n - type: f1\n value: 93.63333333333334\n - type: precision\n value: 93.01666666666665\n - type: recall\n value: 94.89999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tat-eng)\n type: mteb/tatoeba-bitext-mining\n config: tat-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 77.9\n - type: f1\n value: 73.64833333333334\n - type: precision\n value: 71.90282106782105\n - type: recall\n value: 77.9\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (oci-eng)\n type: mteb/tatoeba-bitext-mining\n config: oci-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 59.4\n - type: f1\n value: 54.90521367521367\n - type: precision\n value: 53.432840025471606\n - type: recall\n value: 59.4\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (pol-eng)\n type: mteb/tatoeba-bitext-mining\n config: pol-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97.39999999999999\n - type: f1\n value: 96.6\n - type: precision\n value: 96.2\n - type: recall\n value: 97.39999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (war-eng)\n type: mteb/tatoeba-bitext-mining\n config: war-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 67.2\n - type: f1\n value: 62.25926129426129\n - type: precision\n value: 60.408376623376626\n - type: recall\n value: 67.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (aze-eng)\n type: mteb/tatoeba-bitext-mining\n config: aze-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 90.2\n - type: f1\n value: 87.60666666666667\n - type: precision\n value: 86.45277777777778\n - type: recall\n value: 90.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (vie-eng)\n type: mteb/tatoeba-bitext-mining\n config: vie-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 97.7\n - type: f1\n value: 97\n - type: precision\n value: 96.65\n - type: recall\n value: 97.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (nno-eng)\n type: mteb/tatoeba-bitext-mining\n config: nno-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 93.2\n - type: f1\n value: 91.39746031746031\n - type: precision\n value: 90.6125\n - type: recall\n value: 93.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (cha-eng)\n type: mteb/tatoeba-bitext-mining\n config: cha-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 32.11678832116788\n - type: f1\n value: 27.210415386260234\n - type: precision\n value: 26.20408990846947\n - type: recall\n value: 32.11678832116788\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (mhr-eng)\n type: mteb/tatoeba-bitext-mining\n config: mhr-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 8.5\n - type: f1\n value: 6.787319277832475\n - type: precision\n value: 6.3452094433344435\n - type: recall\n value: 8.5\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (dan-eng)\n type: mteb/tatoeba-bitext-mining\n config: dan-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.1\n - type: f1\n value: 95.08\n - type: precision\n value: 94.61666666666667\n - type: recall\n value: 96.1\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ell-eng)\n type: mteb/tatoeba-bitext-mining\n config: ell-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 95.3\n - type: f1\n value: 93.88333333333333\n - type: precision\n value: 93.18333333333332\n - type: recall\n value: 95.3\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (amh-eng)\n type: mteb/tatoeba-bitext-mining\n config: amh-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 85.11904761904762\n - type: f1\n value: 80.69444444444444\n - type: precision\n value: 78.72023809523809\n - type: recall\n value: 85.11904761904762\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (pam-eng)\n type: mteb/tatoeba-bitext-mining\n config: pam-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 11.1\n - type: f1\n value: 9.276381801735853\n - type: precision\n value: 8.798174603174601\n - type: recall\n value: 11.1\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (hsb-eng)\n type: mteb/tatoeba-bitext-mining\n config: hsb-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 63.56107660455487\n - type: f1\n value: 58.70433569191332\n - type: precision\n value: 56.896926581464015\n - type: recall\n value: 63.56107660455487\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (srp-eng)\n type: mteb/tatoeba-bitext-mining\n config: srp-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.69999999999999\n - type: f1\n value: 93.10000000000001\n - type: precision\n value: 92.35\n - type: recall\n value: 94.69999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (epo-eng)\n type: mteb/tatoeba-bitext-mining\n config: epo-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.8\n - type: f1\n value: 96.01222222222222\n - type: precision\n value: 95.67083333333332\n - type: recall\n value: 96.8\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (kzj-eng)\n type: mteb/tatoeba-bitext-mining\n config: kzj-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 9.2\n - type: f1\n value: 7.911555250305249\n - type: precision\n value: 7.631246556216846\n - type: recall\n value: 9.2\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (awa-eng)\n type: mteb/tatoeba-bitext-mining\n config: awa-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 77.48917748917748\n - type: f1\n value: 72.27375798804371\n - type: precision\n value: 70.14430014430013\n - type: recall\n value: 77.48917748917748\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (fao-eng)\n type: mteb/tatoeba-bitext-mining\n config: fao-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 77.09923664122137\n - type: f1\n value: 72.61541257724463\n - type: precision\n value: 70.8998380754106\n - type: recall\n value: 77.09923664122137\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (mal-eng)\n type: mteb/tatoeba-bitext-mining\n config: mal-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 98.2532751091703\n - type: f1\n value: 97.69529354682193\n - type: precision\n value: 97.42843279961184\n - type: recall\n value: 98.2532751091703\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ile-eng)\n type: mteb/tatoeba-bitext-mining\n config: ile-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 82.8\n - type: f1\n value: 79.14672619047619\n - type: precision\n value: 77.59489247311828\n - type: recall\n value: 82.8\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (bos-eng)\n type: mteb/tatoeba-bitext-mining\n config: bos-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.35028248587571\n - type: f1\n value: 92.86252354048965\n - type: precision\n value: 92.2080979284369\n - type: recall\n value: 94.35028248587571\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (cor-eng)\n type: mteb/tatoeba-bitext-mining\n config: cor-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 8.5\n - type: f1\n value: 6.282429263935621\n - type: precision\n value: 5.783274240739785\n - type: recall\n value: 8.5\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (cat-eng)\n type: mteb/tatoeba-bitext-mining\n config: cat-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 92.7\n - type: f1\n value: 91.025\n - type: precision\n value: 90.30428571428571\n - type: recall\n value: 92.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (eus-eng)\n type: mteb/tatoeba-bitext-mining\n config: eus-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 81\n - type: f1\n value: 77.8232380952381\n - type: precision\n value: 76.60194444444444\n - type: recall\n value: 81\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (yue-eng)\n type: mteb/tatoeba-bitext-mining\n config: yue-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 91\n - type: f1\n value: 88.70857142857142\n - type: precision\n value: 87.7\n - type: recall\n value: 91\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (swe-eng)\n type: mteb/tatoeba-bitext-mining\n config: swe-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.39999999999999\n - type: f1\n value: 95.3\n - type: precision\n value: 94.76666666666667\n - type: recall\n value: 96.39999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (dtp-eng)\n type: mteb/tatoeba-bitext-mining\n config: dtp-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 8.1\n - type: f1\n value: 7.001008218834307\n - type: precision\n value: 6.708329562594269\n - type: recall\n value: 8.1\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (kat-eng)\n type: mteb/tatoeba-bitext-mining\n config: kat-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 87.1313672922252\n - type: f1\n value: 84.09070598748882\n - type: precision\n value: 82.79171454104429\n - type: recall\n value: 87.1313672922252\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (jpn-eng)\n type: mteb/tatoeba-bitext-mining\n config: jpn-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.39999999999999\n - type: f1\n value: 95.28333333333333\n - type: precision\n value: 94.73333333333332\n - type: recall\n value: 96.39999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (csb-eng)\n type: mteb/tatoeba-bitext-mining\n config: csb-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 42.29249011857708\n - type: f1\n value: 36.981018542283365\n - type: precision\n value: 35.415877813576024\n - type: recall\n value: 42.29249011857708\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (xho-eng)\n type: mteb/tatoeba-bitext-mining\n config: xho-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 83.80281690140845\n - type: f1\n value: 80.86854460093896\n - type: precision\n value: 79.60093896713614\n - type: recall\n value: 83.80281690140845\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (orv-eng)\n type: mteb/tatoeba-bitext-mining\n config: orv-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 45.26946107784431\n - type: f1\n value: 39.80235464678088\n - type: precision\n value: 38.14342660001342\n - type: recall\n value: 45.26946107784431\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ind-eng)\n type: mteb/tatoeba-bitext-mining\n config: ind-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.3\n - type: f1\n value: 92.9\n - type: precision\n value: 92.26666666666668\n - type: recall\n value: 94.3\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tuk-eng)\n type: mteb/tatoeba-bitext-mining\n config: tuk-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 37.93103448275862\n - type: f1\n value: 33.15192743764172\n - type: precision\n value: 31.57456528146183\n - type: recall\n value: 37.93103448275862\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (max-eng)\n type: mteb/tatoeba-bitext-mining\n config: max-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 69.01408450704226\n - type: f1\n value: 63.41549295774648\n - type: precision\n value: 61.342778895595806\n - type: recall\n value: 69.01408450704226\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (swh-eng)\n type: mteb/tatoeba-bitext-mining\n config: swh-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 76.66666666666667\n - type: f1\n value: 71.60705960705961\n - type: precision\n value: 69.60683760683762\n - type: recall\n value: 76.66666666666667\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (hin-eng)\n type: mteb/tatoeba-bitext-mining\n config: hin-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 95.8\n - type: f1\n value: 94.48333333333333\n - type: precision\n value: 93.83333333333333\n - type: recall\n value: 95.8\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (dsb-eng)\n type: mteb/tatoeba-bitext-mining\n config: dsb-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 52.81837160751566\n - type: f1\n value: 48.435977731384824\n - type: precision\n value: 47.11291973845539\n - type: recall\n value: 52.81837160751566\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ber-eng)\n type: mteb/tatoeba-bitext-mining\n config: ber-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 44.9\n - type: f1\n value: 38.88962621607783\n - type: precision\n value: 36.95936507936508\n - type: recall\n value: 44.9\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tam-eng)\n type: mteb/tatoeba-bitext-mining\n config: tam-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 90.55374592833876\n - type: f1\n value: 88.22553125484721\n - type: precision\n value: 87.26927252985884\n - type: recall\n value: 90.55374592833876\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (slk-eng)\n type: mteb/tatoeba-bitext-mining\n config: slk-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 94.6\n - type: f1\n value: 93.13333333333333\n - type: precision\n value: 92.45333333333333\n - type: recall\n value: 94.6\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tgl-eng)\n type: mteb/tatoeba-bitext-mining\n config: tgl-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 93.7\n - type: f1\n value: 91.99666666666667\n - type: precision\n value: 91.26666666666668\n - type: recall\n value: 93.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ast-eng)\n type: mteb/tatoeba-bitext-mining\n config: ast-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 85.03937007874016\n - type: f1\n value: 81.75853018372703\n - type: precision\n value: 80.34120734908137\n - type: recall\n value: 85.03937007874016\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (mkd-eng)\n type: mteb/tatoeba-bitext-mining\n config: mkd-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 88.3\n - type: f1\n value: 85.5\n - type: precision\n value: 84.25833333333334\n - type: recall\n value: 88.3\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (khm-eng)\n type: mteb/tatoeba-bitext-mining\n config: khm-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 65.51246537396122\n - type: f1\n value: 60.02297410192148\n - type: precision\n value: 58.133467727289236\n - type: recall\n value: 65.51246537396122\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ces-eng)\n type: mteb/tatoeba-bitext-mining\n config: ces-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96\n - type: f1\n value: 94.89\n - type: precision\n value: 94.39166666666667\n - type: recall\n value: 96\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tzl-eng)\n type: mteb/tatoeba-bitext-mining\n config: tzl-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 57.692307692307686\n - type: f1\n value: 53.162393162393165\n - type: precision\n value: 51.70673076923077\n - type: recall\n value: 57.692307692307686\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (urd-eng)\n type: mteb/tatoeba-bitext-mining\n config: urd-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 91.60000000000001\n - type: f1\n value: 89.21190476190475\n - type: precision\n value: 88.08666666666667\n - type: recall\n value: 91.60000000000001\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (ara-eng)\n type: mteb/tatoeba-bitext-mining\n config: ara-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 88\n - type: f1\n value: 85.47\n - type: precision\n value: 84.43266233766234\n - type: recall\n value: 88\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (kor-eng)\n type: mteb/tatoeba-bitext-mining\n config: kor-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 92.7\n - type: f1\n value: 90.64999999999999\n - type: precision\n value: 89.68333333333332\n - type: recall\n value: 92.7\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (yid-eng)\n type: mteb/tatoeba-bitext-mining\n config: yid-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 80.30660377358491\n - type: f1\n value: 76.33044137466307\n - type: precision\n value: 74.78970125786164\n - type: recall\n value: 80.30660377358491\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (fin-eng)\n type: mteb/tatoeba-bitext-mining\n config: fin-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.39999999999999\n - type: f1\n value: 95.44\n - type: precision\n value: 94.99166666666666\n - type: recall\n value: 96.39999999999999\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (tha-eng)\n type: mteb/tatoeba-bitext-mining\n config: tha-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 96.53284671532847\n - type: f1\n value: 95.37712895377129\n - type: precision\n value: 94.7992700729927\n - type: recall\n value: 96.53284671532847\n - task:\n type: BitextMining\n dataset:\n name: MTEB Tatoeba (wuu-eng)\n type: mteb/tatoeba-bitext-mining\n config: wuu-eng\n split: test\n revision: 9080400076fbadbb4c4dcb136ff4eddc40b42553\n metrics:\n - type: accuracy\n value: 89\n - type: f1\n value: 86.23190476190476\n - type: precision\n value: 85.035\n - type: recall\n value: 89\n - task:\n type: Retrieval\n dataset:\n name: MTEB Touche2020\n type: webis-touche2020\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 2.585\n - type: map_at_10\n value: 9.012\n - type: map_at_100\n value: 14.027000000000001\n - type: map_at_1000\n value: 15.565000000000001\n - type: map_at_3\n value: 5.032\n - type: map_at_5\n value: 6.657\n - type: mrr_at_1\n value: 28.571\n - type: mrr_at_10\n value: 45.377\n - type: mrr_at_100\n value: 46.119\n - type: mrr_at_1000\n value: 46.127\n - type: mrr_at_3\n value: 41.156\n - type: mrr_at_5\n value: 42.585\n - type: ndcg_at_1\n value: 27.551\n - type: ndcg_at_10\n value: 23.395\n - type: ndcg_at_100\n value: 33.342\n - type: ndcg_at_1000\n value: 45.523\n - type: ndcg_at_3\n value: 25.158\n - type: ndcg_at_5\n value: 23.427\n - type: precision_at_1\n value: 28.571\n - type: precision_at_10\n value: 21.429000000000002\n - type: precision_at_100\n value: 6.714\n - type: precision_at_1000\n value: 1.473\n - type: precision_at_3\n value: 27.211000000000002\n - type: precision_at_5\n value: 24.490000000000002\n - type: recall_at_1\n value: 2.585\n - type: recall_at_10\n value: 15.418999999999999\n - type: recall_at_100\n value: 42.485\n - type: recall_at_1000\n value: 79.536\n - type: recall_at_3\n value: 6.239999999999999\n - type: recall_at_5\n value: 8.996\n - task:\n type: Classification\n dataset:\n name: MTEB ToxicConversationsClassification\n type: mteb/toxic_conversations_50k\n config: default\n split: test\n revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c\n metrics:\n - type: accuracy\n value: 71.3234\n - type: ap\n value: 14.361688653847423\n - type: f1\n value: 54.819068624319044\n - task:\n type: Classification\n dataset:\n name: MTEB TweetSentimentExtractionClassification\n type: mteb/tweet_sentiment_extraction\n config: default\n split: test\n revision: d604517c81ca91fe16a244d1248fc021f9ecee7a\n metrics:\n - type: accuracy\n value: 61.97792869269949\n - type: f1\n value: 62.28965628513728\n - task:\n type: Clustering\n dataset:\n name: MTEB TwentyNewsgroupsClustering\n type: mteb/twentynewsgroups-clustering\n config: default\n split: test\n revision: 6125ec4e24fa026cec8a478383ee943acfbd5449\n metrics:\n - type: v_measure\n value: 38.90540145385218\n - task:\n type: PairClassification\n dataset:\n name: MTEB TwitterSemEval2015\n type: mteb/twittersemeval2015-pairclassification\n config: default\n split: test\n revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1\n metrics:\n - type: cos_sim_accuracy\n value: 86.53513739047506\n - type: cos_sim_ap\n value: 75.27741586677557\n - type: cos_sim_f1\n value: 69.18792902473774\n - type: cos_sim_precision\n value: 67.94708725515136\n - type: cos_sim_recall\n value: 70.47493403693932\n - type: dot_accuracy\n value: 84.7052512368123\n - type: dot_ap\n value: 69.36075482849378\n - type: dot_f1\n value: 64.44688376631296\n - type: dot_precision\n value: 59.92288500793831\n - type: dot_recall\n value: 69.70976253298153\n - type: euclidean_accuracy\n value: 86.60666388508076\n - type: euclidean_ap\n value: 75.47512772621097\n - type: euclidean_f1\n value: 69.413872536473\n - type: euclidean_precision\n value: 67.39562624254472\n - type: euclidean_recall\n value: 71.55672823218997\n - type: manhattan_accuracy\n value: 86.52917684925792\n - type: manhattan_ap\n value: 75.34000110496703\n - type: manhattan_f1\n value: 69.28489190226429\n - type: manhattan_precision\n value: 67.24608889992551\n - type: manhattan_recall\n value: 71.45118733509234\n - type: max_accuracy\n value: 86.60666388508076\n - type: max_ap\n value: 75.47512772621097\n - type: max_f1\n value: 69.413872536473\n - task:\n type: PairClassification\n dataset:\n name: MTEB TwitterURLCorpus\n type: mteb/twitterurlcorpus-pairclassification\n config: default\n split: test\n revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf\n metrics:\n - type: cos_sim_accuracy\n value: 89.01695967710637\n - type: cos_sim_ap\n value: 85.8298270742901\n - type: cos_sim_f1\n value: 78.46988128389272\n - type: cos_sim_precision\n value: 74.86017897091722\n - type: cos_sim_recall\n value: 82.44533415460425\n - type: dot_accuracy\n value: 88.19420188613343\n - type: dot_ap\n value: 83.82679165901324\n - type: dot_f1\n value: 76.55833777304208\n - type: dot_precision\n value: 75.6884875846501\n - type: dot_recall\n value: 77.44841392054204\n - type: euclidean_accuracy\n value: 89.03054294252338\n - type: euclidean_ap\n value: 85.89089555185325\n - type: euclidean_f1\n value: 78.62997658079624\n - type: euclidean_precision\n value: 74.92329149232914\n - type: euclidean_recall\n value: 82.72251308900523\n - type: manhattan_accuracy\n value: 89.0266620095471\n - type: manhattan_ap\n value: 85.86458997929147\n - type: manhattan_f1\n value: 78.50685331000291\n - type: manhattan_precision\n value: 74.5499861534201\n - type: manhattan_recall\n value: 82.90729904527257\n - type: max_accuracy\n value: 89.03054294252338\n - type: max_ap\n value: 85.89089555185325\n - type: max_f1\n value: 78.62997658079624\n---\n\n# CyberOptic/multilingual-e5-large-Q8_0-GGUF\nThis model was converted to GGUF format from [`intfloat/multilingual-e5-large`](https://huggingface.co/intfloat/multilingual-e5-large) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.\nRefer to the [original model card](https://huggingface.co/intfloat/multilingual-e5-large) for more details on the model.\n\n## Use with llama.cpp\nInstall llama.cpp through brew (works on Mac and Linux)\n\n```bash\nbrew install llama.cpp\n\n```\nInvoke the llama.cpp server or the CLI.\n\n### CLI:\n```bash\nllama-cli --hf-repo CyberOptic/multilingual-e5-large-Q8_0-GGUF --hf-file multilingual-e5-large-q8_0.gguf -p \"The meaning to life and the universe is\"\n```\n\n### Server:\n```bash\nllama-server --hf-repo CyberOptic/multilingual-e5-large-Q8_0-GGUF --hf-file multilingual-e5-large-q8_0.gguf -c 2048\n```\n\nNote: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.\n\nStep 1: Clone llama.cpp from GitHub.\n```\ngit clone https://github.com/ggerganov/llama.cpp\n```\n\nStep 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).\n```\ncd llama.cpp && LLAMA_CURL=1 make\n```\n\nStep 3: Run inference through the main binary.\n```\n./llama-cli --hf-repo CyberOptic/multilingual-e5-large-Q8_0-GGUF --hf-file multilingual-e5-large-q8_0.gguf -p \"The meaning to life and the universe is\"\n```\nor \n```\n./llama-server --hf-repo CyberOptic/multilingual-e5-large-Q8_0-GGUF --hf-file multilingual-e5-large-q8_0.gguf -c 2048\n```\n"},"matched_task":{"kind":"list like","value":["SUMMARIZATION"],"string":"[\n \"SUMMARIZATION\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["BIOSSES","SCIFACT"],"string":"[\n \"BIOSSES\",\n \"SCIFACT\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"Non_BioNLP"}}},{"rowIdx":294,"cells":{"id":{"kind":"string","value":"LoneStriker/OpenBioLLM-Llama3-8B-5.0bpw-h6-exl2"},"author":{"kind":"string","value":"LoneStriker"},"task_category":{"kind":"string","value":"text-generation"},"tags":{"kind":"list like","value":["transformers","pytorch","llama","text-generation","llama-3","Mixtral","instruct","finetune","chatml","DPO","RLHF","gpt4","distillation","en","arxiv:2305.18290","arxiv:2303.13375","arxiv:2212.13138","arxiv:2305.09617","arxiv:2402.07023","base_model:meta-llama/Meta-Llama-3-8B","base_model:quantized:meta-llama/Meta-Llama-3-8B","license:llama3","autotrain_compatible","text-generation-inference","endpoints_compatible","5-bit","exl2","region:us"],"string":"[\n \"transformers\",\n \"pytorch\",\n \"llama\",\n \"text-generation\",\n \"llama-3\",\n \"Mixtral\",\n \"instruct\",\n \"finetune\",\n \"chatml\",\n \"DPO\",\n \"RLHF\",\n \"gpt4\",\n \"distillation\",\n \"en\",\n \"arxiv:2305.18290\",\n \"arxiv:2303.13375\",\n \"arxiv:2212.13138\",\n \"arxiv:2305.09617\",\n \"arxiv:2402.07023\",\n \"base_model:meta-llama/Meta-Llama-3-8B\",\n \"base_model:quantized:meta-llama/Meta-Llama-3-8B\",\n \"license:llama3\",\n \"autotrain_compatible\",\n \"text-generation-inference\",\n \"endpoints_compatible\",\n \"5-bit\",\n \"exl2\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1714,"string":"1,714"},"last_modified":{"kind":"number","value":1714,"string":"1,714"},"downloads":{"kind":"number","value":8,"string":"8"},"likes":{"kind":"number","value":0,"string":"0"},"README":{"kind":"string","value":"---\nbase_model: meta-llama/Meta-Llama-3-8B\nlanguage:\n- en\nlicense: llama3\ntags:\n- llama-3\n- llama\n- Mixtral\n- instruct\n- finetune\n- chatml\n- DPO\n- RLHF\n- gpt4\n- distillation\nwidget:\n- example_title: OpenBioLLM-8B\n messages:\n - role: system\n content: You are an expert and experienced from the healthcare and biomedical\n domain with extensive medical knowledge and practical experience.\n - role: user\n content: How long does it take for newborn jaundice to go away?\n output:\n text: 'Newborn jaundice, also known as neonatal jaundice, is a common condition\n in newborns where the yellowing of the skin and eyes occurs due to an elevated\n level of bilirubin in the blood. Bilirubin is a yellow pigment that forms when\n red blood cells break down. In most cases, newborn jaundice resolves on its\n own without any specific treatment.\n\n The duration of newborn jaundice can vary depending on several factors such\n as the underlying cause, gestational age at birth, and individual variations\n in bilirubin metabolism. Here are some general guidelines:\n\n 1. Physiological jaundice: This is the most common type of newborn jaundice\n and usually appears within 24-72 hours after birth. It tends to peak between\n the second and fifth day of life and gradually improves over the next week or\n two. By the time the baby is one week old, the jaundice should have mostly resolved.\n 2. Breast milk jaundice: This type of jaundice occurs in breastfed babies and\n may appear later than physiological jaundice, typically between the fifth and\n fourteenth day of life. It tends to persist for a longer duration but usually\n resolves within six weeks after birth. 3. Pathological jaundice: This type of\n jaundice is less common and occurs due to an underlying medical condition that\n affects bilirubin metabolism or liver function. The duration of pathological\n jaundice depends on the specific cause and may require treatment.\n\n It''s important for parents to monitor their newborn''s jaundice closely and\n seek medical advice if the jaundice progresses rapidly, becomes severe, or is\n accompanied by other symptoms such as poor feeding, lethargy, or excessive sleepiness.\n In these cases, further evaluation and management may be necessary. Remember\n that each baby is unique, and the timing of jaundice resolution can vary. If\n you have concerns about your newborn''s jaundice, it''s always best to consult\n with a healthcare professional for personalized advice and guidance.'\nmodel-index:\n- name: OpenBioLLM-8B\n results: []\n---\n\n\n<div align=\"center\">\n<img width=\"260px\" src=\"https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/BrQCb95lmEIFz79QAmoNA.png\"></div>\n\n\n\n\n\n\n\n<div align=\"center\">\n \n <h1>Advancing Open-source Large Language Models in Medical Domain</h1>\n</div>\n\n<p align=\"center\" style=\"margin-top: 0px;\">\n <a href=\"https://colab.research.google.com/drive/1F5oV20InEYeAJGmBwYF9NM_QhLmjBkKJ?usp=sharing\">\n <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"OpenChat Logo\" style=\"width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 10px; margin-top: 0px; margin-bottom: 0px;\"/>\n <span class=\"link-text\" style=\" margin-right: 5px;\">Online Demo</span>\n </a> |\n <a href=\"https://github.com/openlifescience-ai\">\n <img src=\"https://github.githubassets.com/assets/GitHub-Mark-ea2971cee799.png\" alt=\"GitHub Logo\" style=\"width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 5px; margin-top: 0px; margin-bottom: 0px;\"/>\n <span class=\"link-text\" style=\" margin-right: 5px;\">GitHub</span>\n </a> |\n <a href=\"#\">\n <img src=\"https://github.com/alpayariyak/openchat/blob/master/assets/arxiv-logomark-small-square-border.png?raw=true\" alt=\"ArXiv Logo\" style=\"width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 5px; margin-top: 0px; margin-bottom: 0px;\"/>\n <span class=\"link-text\" style=\"margin-right: 5px;\">Paper</span>\n </a> |\n <a href=\"https://discord.gg/A5Fjf5zC69\">\n <img src=\"https://cloud.githubusercontent.com/assets/6291467/26705903/96c2d66e-477c-11e7-9f4e-f3c0efe96c9a.png\" alt=\"Discord Logo\" style=\"width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 5px; margin-top: 0px; margin-bottom: 0px;\"/>\n <span class=\"link-text\">Discord</span>\n </a>\n</p>\n\n\n\nIntroducing OpenBioLLM-8B: A State-of-the-Art Open Source Biomedical Large Language Model\n\n\nOpenBioLLM-8B is an advanced open source language model designed specifically for the biomedical domain. Developed by Saama AI Labs, this model leverages cutting-edge techniques to achieve state-of-the-art performance on a wide range of biomedical tasks.\n\n🏥 **Biomedical Specialization**: OpenBioLLM-8B is tailored for the unique language and knowledge requirements of the medical and life sciences fields. It was fine-tuned on a vast corpus of high-quality biomedical data, enabling it to understand and generate text with domain-specific accuracy and fluency.\n\n🎓 **Superior Performance**: With 8 billion parameters, OpenBioLLM-8B outperforms other open source biomedical language models of similar scale. It has also demonstrated better results compared to larger proprietary & open-source models like GPT-3.5 and Meditron-70B on biomedical benchmarks.\n\n🧠 **Advanced Training Techniques**: OpenBioLLM-8B builds upon the powerful foundations of the **Meta-Llama-3-8B** and [Meta-Llama-3-8B](meta-llama/Meta-Llama-3-8B) models. It incorporates the DPO dataset and fine-tuning recipe along with a custom diverse medical instruction dataset. Key components of the training pipeline include:\n\n<div align=\"center\">\n<img width=\"1200px\" src=\"https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/oPchsJsEpQoGcGXVbh7YS.png\">\n</div>\n\n- **Policy Optimization**: [Direct Preference Optimization: Your Language Model is Secretly a Reward Model (DPO)](https://arxiv.org/abs/2305.18290)\n- **Ranking Dataset**: [berkeley-nest/Nectar](https://huggingface.co/datasets/berkeley-nest/Nectar)\n- **Fine-tuning dataset**: Custom Medical Instruct dataset (We plan to release a sample training dataset in our upcoming paper; please stay updated)\n\nThis combination of cutting-edge techniques enables OpenBioLLM-8B to align with key capabilities and preferences for biomedical applications.\n\n⚙️ **Release Details**:\n\n- **Model Size**: 8 billion parameters\n- **Quantization**: Optimized quantized versions available [Here](https://huggingface.co/aaditya/OpenBioLLM-8B-GGUF)\n- **Language(s) (NLP):** en\n- **Developed By**: [Ankit Pal (Aaditya Ura)](https://aadityaura.github.io/) from Saama AI Labs \n- **License:** Meta-Llama License \n- **Fine-tuned from models:** [meta-llama/Meta-Llama-3-8B](meta-llama/Meta-Llama-3-8B)\n- **Resources for more information:**\n - Paper: Coming soon\n\nThe model can be fine-tuned for more specialized tasks and datasets as needed.\n\nOpenBioLLM-8B represents an important step forward in democratizing advanced language AI for the biomedical community. By leveraging state-of-the-art architectures and training techniques from leading open source efforts like Llama-3, we have created a powerful tool to accelerate innovation and discovery in healthcare and the life sciences.\n\nWe are excited to share OpenBioLLM-8B with researchers and developers around the world.\n\n\n### Use with transformers\n\n**Important: Please use the exact chat template provided by Llama-3 instruct version. Otherwise there will be a degradation in the performance. The model output can be verbose in rare cases. Please consider setting temperature = 0 to make this happen less.**\n\nSee the snippet below for usage with Transformers:\n\n```python\nimport transformers\nimport torch\n\nmodel_id = \"aaditya/OpenBioLLM-Llama3-8B\"\n\npipeline = transformers.pipeline(\n \"text-generation\",\n model=model_id,\n model_kwargs={\"torch_dtype\": torch.bfloat16},\n device=\"auto\",\n)\n\nmessages = [\n {\"role\": \"system\", \"content\": \"You are an expert and experienced from the healthcare and biomedical domain with extensive medical knowledge and practical experience. Your name is OpenBioLLM, and you were developed by Saama AI Labs. who's willing to help answer the user's query with explanation. In your explanation, leverage your deep medical expertise such as relevant anatomical structures, physiological processes, diagnostic criteria, treatment guidelines, or other pertinent medical concepts. Use precise medical terminology while still aiming to make the explanation clear and accessible to a general audience.\"},\n {\"role\": \"user\", \"content\": \"How can i split a 3mg or 4mg waefin pill so i can get a 2.5mg pill?\"},\n]\n\nprompt = pipeline.tokenizer.apply_chat_template(\n\t\tmessages, \n\t\ttokenize=False, \n\t\tadd_generation_prompt=True\n)\n\nterminators = [\n pipeline.tokenizer.eos_token_id,\n pipeline.tokenizer.convert_tokens_to_ids(\"<|eot_id|>\")\n]\n\noutputs = pipeline(\n prompt,\n max_new_tokens=256,\n eos_token_id=terminators,\n do_sample=True,\n temperature=0.0,\n top_p=0.9,\n)\nprint(outputs[0][\"generated_text\"][len(prompt):])\n```\n\n## **Training procedure**\n\n### **Training hyperparameters**\n\n<details>\n <summary>Click to see details</summary>\n\n- learning_rate: 0.0002\n- lr_scheduler: cosine\n- train_batch_size: 12\n- eval_batch_size: 8\n- GPU: H100 80GB SXM5\n- num_devices: 1\n- optimizer: adamw_bnb_8bit\n- lr_scheduler_warmup_steps: 100\n- num_epochs: 4\n</details>\n\n \n### **Peft hyperparameters**\n\n<details>\n <summary>Click to see details</summary>\n\n- adapter: qlora\n- lora_r: 128\n- lora_alpha: 256\n- lora_dropout: 0.05\n- lora_target_linear: true\n \n-lora_target_modules:\n - q_proj\n - v_proj\n - k_proj\n - o_proj\n - gate_proj\n - down_proj\n - up_proj\n</details>\n\n\n\n### **Training results**\n\n### **Framework versions**\n\n- Transformers 4.39.3\n- Pytorch 2.1.2+cu121\n- Datasets 2.18.0\n- Tokenizers 0.15.1\n- Axolotl\n- Lm harness for evaluation\n\n\n# Benchmark Results\n\n🔥 OpenBioLLM-8B demonstrates superior performance compared to larger models, such as GPT-3.5, Meditron-70B across 9 diverse biomedical datasets, achieving state-of-the-art results with an average score of 72.50%, despite having a significantly smaller parameter count. The model's strong performance in domain-specific tasks, such as Clinical KG, Medical Genetics, and PubMedQA, highlights its ability to effectively capture and apply biomedical knowledge.\n\n🚨 The GPT-4, Med-PaLM-1, and Med-PaLM-2 results are taken from their official papers. Since Med-PaLM doesn't provide zero-shot accuracy, we are using 5-shot accuracy from their paper for comparison. All results presented are in the zero-shot setting, except for Med-PaLM-2 and Med-PaLM-1, which use 5-shot accuracy.\n\n| | Clinical KG | Medical Genetics | Anatomy | Pro Medicine | College Biology | College Medicine | MedQA 4 opts | PubMedQA | MedMCQA | Avg |\n|--------------------|-------------|------------------|---------|--------------|-----------------|------------------|--------------|----------|---------|-------|\n| **OpenBioLLM-70B** | **92.93** | **93.197** | **83.904** | 93.75 | 93.827 | **85.749** | 78.162 | 78.97 | **74.014** | **86.05588** |\n| Med-PaLM-2 (5-shot) | 88.3 | 90 | 77.8 | **95.2** | 94.4 | 80.9 | **79.7** | **79.2** | 71.3 | 84.08 |\n| **GPT-4** | 86.04 | 91 | 80 | 93.01 | **95.14** | 76.88 | 78.87 | 75.2 | 69.52 | 82.85 |\n| Med-PaLM-1 (Flan-PaLM, 5-shot) | 80.4 | 75 | 63.7 | 83.8 | 88.9 | 76.3 | 67.6 | 79 | 57.6 | 74.7 |\n| **OpenBioLLM-8B** | 76.101 | 86.1 | 69.829 | 78.21 | 84.213 | 68.042 | 58.993 | 74.12 | 56.913 | 72.502 |\n| Gemini-1.0 | 76.7 | 75.8 | 66.7 | 77.7 | 88 | 69.2 | 58 | 70.7 | 54.3 | 70.79 |\n| GPT-3.5 Turbo 1106 | 74.71 | 74 | 72.79 | 72.79 | 72.91 | 64.73 | 57.71 | 72.66 | 53.79 | 66 |\n| Meditron-70B | 66.79 | 69 | 53.33 | 71.69 | 76.38 | 63 | 57.1 | 76.6 | 46.85 | 64.52 |\n| gemma-7b | 69.81 | 70 | 59.26 | 66.18 | 79.86 | 60.12 | 47.21 | 76.2 | 48.96 | 64.18 |\n| Mistral-7B-v0.1 | 68.68 | 71 | 55.56 | 68.38 | 68.06 | 59.54 | 50.82 | 75.4 | 48.2 | 62.85 |\n| Apollo-7B | 62.26 | 72 | 61.48 | 69.12 | 70.83 | 55.49 | 55.22 | 39.8 | 53.77 | 60 |\n| MedAlpaca-7b | 57.36 | 69 | 57.04 | 67.28 | 65.28 | 54.34 | 41.71 | 72.8 | 37.51 | 58.03 |\n| BioMistral-7B | 59.9 | 64 | 56.5 | 60.4 | 59 | 54.7 | 50.6 | 77.5 | 48.1 | 57.3 |\n| AlpaCare-llama2-7b | 49.81 | 49 | 45.92 | 33.82 | 50 | 43.35 | 29.77 | 72.2 | 34.42 | 45.36 |\n| ClinicalGPT | 30.56 | 27 | 30.37 | 19.48 | 25 | 24.27 | 26.08 | 63.8 | 28.18 | 30.52 |\n\n<div align=\"center\">\n<img width=\"1600px\" src=\"https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/_SzdcJSBjZyo8RS1bTEkP.png\">\n</div>\n\n## Detailed Medical Subjectwise accuracy\n\n\n\n\n# Use Cases & Examples\n\n🚨 **Below results are from the quantized version of OpenBioLLM-70B**\n\n\n# Summarize Clinical Notes\n\nOpenBioLLM-70B can efficiently analyze and summarize complex clinical notes, EHR data, and discharge summaries, extracting key information and generating concise, structured summaries\n\n\n\n\n# Answer Medical Questions\n\nOpenBioLLM-70B can provide answers to a wide range of medical questions.\n\n\n\n\n\n<details>\n <summary>Click to see details</summary>\n\n\n\n\n\n\n</details>\n\n# Clinical Entity Recognition\n\nOpenBioLLM-70B can perform advanced clinical entity recognition by identifying and extracting key medical concepts, such as diseases, symptoms, medications, procedures, and anatomical structures, from unstructured clinical text. By leveraging its deep understanding of medical terminology and context, the model can accurately annotate and categorize clinical entities, enabling more efficient information retrieval, data analysis, and knowledge discovery from electronic health records, research articles, and other biomedical text sources. This capability can support various downstream applications, such as clinical decision support, pharmacovigilance, and medical research.\n\n\n\n\n\n\n# Biomarkers Extraction\n\n\n\n\n\n# Classification\n\nOpenBioLLM-70B can perform various biomedical classification tasks, such as disease prediction, sentiment analysis, medical document categorization\n\n\n\n# De-Identification\n\nOpenBioLLM-70B can detect and remove personally identifiable information (PII) from medical records, ensuring patient privacy and compliance with data protection regulations like HIPAA.\n\n\n\n\n\n**Advisory Notice!** \n\nWhile OpenBioLLM-70B & 8B leverages high-quality data sources, its outputs may still contain inaccuracies, biases, or misalignments that could pose risks if relied upon for medical decision-making without further testing and refinement. The model's performance has not yet been rigorously evaluated in randomized controlled trials or real-world healthcare environments.\n\nTherefore, we strongly advise against using OpenBioLLM-70B & 8B for any direct patient care, clinical decision support, or other professional medical purposes at this time. Its use should be limited to research, development, and exploratory applications by qualified individuals who understand its limitations.\nOpenBioLLM-70B & 8B are intended solely as a research tool to assist healthcare professionals and should never be considered a replacement for the professional judgment and expertise of a qualified medical doctor.\n\nAppropriately adapting and validating OpenBioLLM-70B & 8B for specific medical use cases would require significant additional work, potentially including:\n\n- Thorough testing and evaluation in relevant clinical scenarios\n- Alignment with evidence-based guidelines and best practices\n- Mitigation of potential biases and failure modes\n- Integration with human oversight and interpretation\n- Compliance with regulatory and ethical standards\n\nAlways consult a qualified healthcare provider for personal medical needs.\n\n\n\n# Citation\n\nIf you find OpenBioLLM-70B & 8B useful in your work, please cite the model as follows:\n\n```\n@misc{OpenBioLLMs,\n author = {Ankit Pal, Malaikannan Sankarasubbu},\n title = {OpenBioLLMs: Advancing Open-Source Large Language Models for Healthcare and Life Sciences},\n year = {2024},\n publisher = {Hugging Face},\n journal = {Hugging Face repository},\n howpublished = {\\url{https://huggingface.co/aaditya/OpenBioLLM-Llama3-70B}}\n}\n```\n\nThe accompanying paper is currently in progress and will be released soon.\n\n<div align=\"center\">\n<h2> 💌 Contact </h2>\n</div>\n\nWe look forward to hearing you and collaborating on this exciting project!\n\n**Contributors:**\n- [Ankit Pal (Aaditya Ura)](https://aadityaura.github.io/) [aadityaura at gmail dot com]\n- Saama AI Labs\n- Note: I am looking for a funded PhD opportunity, especially if it fits my Responsible Generative AI, Multimodal LLMs, Geometric Deep Learning, and Healthcare AI skillset.\n\n\n# References\n\nWe thank the [Meta Team](meta-llama/Meta-Llama-3-70B-Instruct) for their amazing models!\n\n\nResult sources\n\n- [1] GPT-4 [Capabilities of GPT-4 on Medical Challenge Problems] (https://arxiv.org/abs/2303.13375) \n- [2] Med-PaLM-1 [Large Language Models Encode Clinical Knowledge](https://arxiv.org/abs/2212.13138)\n- [3] Med-PaLM-2 [Towards Expert-Level Medical Question Answering with Large Language Models](https://arxiv.org/abs/2305.09617)\n- [4] Gemini-1.0 [Gemini Goes to Med School](https://arxiv.org/abs/2402.07023)"},"matched_task":{"kind":"list like","value":["QUESTION_ANSWERING"],"string":"[\n \"QUESTION_ANSWERING\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["MEDQA","PUBMEDQA"],"string":"[\n \"MEDQA\",\n \"PUBMEDQA\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"BioNLP"}}},{"rowIdx":295,"cells":{"id":{"kind":"string","value":"hkunlp/instructor-base"},"author":{"kind":"string","value":"hkunlp"},"task_category":{"kind":"string","value":"sentence-similarity"},"tags":{"kind":"list like","value":["sentence-transformers","pytorch","t5","text-embedding","embeddings","information-retrieval","beir","text-classification","language-model","text-clustering","text-semantic-similarity","text-evaluation","prompt-retrieval","text-reranking","feature-extraction","sentence-similarity","transformers","English","Sentence Similarity","natural_questions","ms_marco","fever","hotpot_qa","mteb","en","arxiv:2212.09741","license:apache-2.0","model-index","autotrain_compatible","text-generation-inference","region:us"],"string":"[\n \"sentence-transformers\",\n \"pytorch\",\n \"t5\",\n \"text-embedding\",\n \"embeddings\",\n \"information-retrieval\",\n \"beir\",\n \"text-classification\",\n \"language-model\",\n \"text-clustering\",\n \"text-semantic-similarity\",\n \"text-evaluation\",\n \"prompt-retrieval\",\n \"text-reranking\",\n \"feature-extraction\",\n \"sentence-similarity\",\n \"transformers\",\n \"English\",\n \"Sentence Similarity\",\n \"natural_questions\",\n \"ms_marco\",\n \"fever\",\n \"hotpot_qa\",\n \"mteb\",\n \"en\",\n \"arxiv:2212.09741\",\n \"license:apache-2.0\",\n \"model-index\",\n \"autotrain_compatible\",\n \"text-generation-inference\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1671,"string":"1,671"},"last_modified":{"kind":"number","value":1674,"string":"1,674"},"downloads":{"kind":"number","value":11563,"string":"11,563"},"likes":{"kind":"number","value":116,"string":"116"},"README":{"kind":"string","value":"---\nlanguage: en\nlicense: apache-2.0\npipeline_tag: sentence-similarity\ntags:\n- text-embedding\n- embeddings\n- information-retrieval\n- beir\n- text-classification\n- language-model\n- text-clustering\n- text-semantic-similarity\n- text-evaluation\n- prompt-retrieval\n- text-reranking\n- sentence-transformers\n- feature-extraction\n- sentence-similarity\n- transformers\n- t5\n- English\n- Sentence Similarity\n- natural_questions\n- ms_marco\n- fever\n- hotpot_qa\n- mteb\ninference: false\nmodel-index:\n- name: final_base_results\n results:\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonCounterfactualClassification (en)\n type: mteb/amazon_counterfactual\n config: en\n split: test\n revision: e8379541af4e31359cca9fbcf4b00f2671dba205\n metrics:\n - type: accuracy\n value: 86.2089552238806\n - type: ap\n value: 55.76273850794966\n - type: f1\n value: 81.26104211414781\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonPolarityClassification\n type: mteb/amazon_polarity\n config: default\n split: test\n revision: e2d317d38cd51312af73b3d32a06d1a08b442046\n metrics:\n - type: accuracy\n value: 88.35995000000001\n - type: ap\n value: 84.18839957309655\n - type: f1\n value: 88.317619250081\n - task:\n type: Classification\n dataset:\n name: MTEB AmazonReviewsClassification (en)\n type: mteb/amazon_reviews_multi\n config: en\n split: test\n revision: 1399c76144fd37290681b995c656ef9b2e06e26d\n metrics:\n - type: accuracy\n value: 44.64\n - type: f1\n value: 42.48663956478136\n - task:\n type: Retrieval\n dataset:\n name: MTEB ArguAna\n type: arguana\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 27.383000000000003\n - type: map_at_10\n value: 43.024\n - type: map_at_100\n value: 44.023\n - type: map_at_1000\n value: 44.025999999999996\n - type: map_at_3\n value: 37.684\n - type: map_at_5\n value: 40.884\n - type: mrr_at_1\n value: 28.094\n - type: mrr_at_10\n value: 43.315\n - type: mrr_at_100\n value: 44.313\n - type: mrr_at_1000\n value: 44.317\n - type: mrr_at_3\n value: 37.862\n - type: mrr_at_5\n value: 41.155\n - type: ndcg_at_1\n value: 27.383000000000003\n - type: ndcg_at_10\n value: 52.032000000000004\n - type: ndcg_at_100\n value: 56.19499999999999\n - type: ndcg_at_1000\n value: 56.272\n - type: ndcg_at_3\n value: 41.166000000000004\n - type: ndcg_at_5\n value: 46.92\n - type: precision_at_1\n value: 27.383000000000003\n - type: precision_at_10\n value: 8.087\n - type: precision_at_100\n value: 0.989\n - type: precision_at_1000\n value: 0.099\n - type: precision_at_3\n value: 17.093\n - type: precision_at_5\n value: 13.044\n - type: recall_at_1\n value: 27.383000000000003\n - type: recall_at_10\n value: 80.868\n - type: recall_at_100\n value: 98.86200000000001\n - type: recall_at_1000\n value: 99.431\n - type: recall_at_3\n value: 51.28\n - type: recall_at_5\n value: 65.22\n - task:\n type: Clustering\n dataset:\n name: MTEB ArxivClusteringP2P\n type: mteb/arxiv-clustering-p2p\n config: default\n split: test\n revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d\n metrics:\n - type: v_measure\n value: 39.68441054431849\n - task:\n type: Clustering\n dataset:\n name: MTEB ArxivClusteringS2S\n type: mteb/arxiv-clustering-s2s\n config: default\n split: test\n revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53\n metrics:\n - type: v_measure\n value: 29.188539728343844\n - task:\n type: Reranking\n dataset:\n name: MTEB AskUbuntuDupQuestions\n type: mteb/askubuntudupquestions-reranking\n config: default\n split: test\n revision: 2000358ca161889fa9c082cb41daa8dcfb161a54\n metrics:\n - type: map\n value: 63.173362687519784\n - type: mrr\n value: 76.18860748362133\n - task:\n type: STS\n dataset:\n name: MTEB BIOSSES\n type: mteb/biosses-sts\n config: default\n split: test\n revision: d3fb88f8f02e40887cd149695127462bbcf29b4a\n metrics:\n - type: cos_sim_spearman\n value: 82.30789953771232\n - task:\n type: Classification\n dataset:\n name: MTEB Banking77Classification\n type: mteb/banking77\n config: default\n split: test\n revision: 0fd18e25b25c072e09e0d92ab615fda904d66300\n metrics:\n - type: accuracy\n value: 77.03571428571428\n - type: f1\n value: 75.87384305045917\n - task:\n type: Clustering\n dataset:\n name: MTEB BiorxivClusteringP2P\n type: mteb/biorxiv-clustering-p2p\n config: default\n split: test\n revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40\n metrics:\n - type: v_measure\n value: 32.98041170516364\n - task:\n type: Clustering\n dataset:\n name: MTEB BiorxivClusteringS2S\n type: mteb/biorxiv-clustering-s2s\n config: default\n split: test\n revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908\n metrics:\n - type: v_measure\n value: 25.71652988451154\n - task:\n type: Retrieval\n dataset:\n name: MTEB CQADupstackAndroidRetrieval\n type: BeIR/cqadupstack\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 33.739999999999995\n - type: map_at_10\n value: 46.197\n - type: map_at_100\n value: 47.814\n - type: map_at_1000\n value: 47.934\n - type: map_at_3\n value: 43.091\n - type: map_at_5\n value: 44.81\n - type: mrr_at_1\n value: 41.059\n - type: mrr_at_10\n value: 52.292\n - type: mrr_at_100\n value: 52.978\n - type: mrr_at_1000\n value: 53.015\n - type: mrr_at_3\n value: 49.976\n - type: mrr_at_5\n value: 51.449999999999996\n - type: ndcg_at_1\n value: 41.059\n - type: ndcg_at_10\n value: 52.608\n - type: ndcg_at_100\n value: 57.965\n - type: ndcg_at_1000\n value: 59.775999999999996\n - type: ndcg_at_3\n value: 48.473\n - type: ndcg_at_5\n value: 50.407999999999994\n - type: precision_at_1\n value: 41.059\n - type: precision_at_10\n value: 9.943\n - type: precision_at_100\n value: 1.6070000000000002\n - type: precision_at_1000\n value: 0.20500000000000002\n - type: precision_at_3\n value: 23.413999999999998\n - type: precision_at_5\n value: 16.481\n - type: recall_at_1\n value: 33.739999999999995\n - type: recall_at_10\n value: 63.888999999999996\n - type: recall_at_100\n value: 85.832\n - type: recall_at_1000\n value: 97.475\n - type: recall_at_3\n value: 51.953\n - type: recall_at_5\n value: 57.498000000000005\n - type: map_at_1\n value: 31.169999999999998\n - type: map_at_10\n value: 41.455\n - type: map_at_100\n value: 42.716\n - type: map_at_1000\n value: 42.847\n - type: map_at_3\n value: 38.568999999999996\n - type: map_at_5\n value: 40.099000000000004\n - type: mrr_at_1\n value: 39.427\n - type: mrr_at_10\n value: 47.818\n - type: mrr_at_100\n value: 48.519\n - type: mrr_at_1000\n value: 48.558\n - type: mrr_at_3\n value: 45.86\n - type: mrr_at_5\n value: 46.936\n - type: ndcg_at_1\n value: 39.427\n - type: ndcg_at_10\n value: 47.181\n - type: ndcg_at_100\n value: 51.737\n - type: ndcg_at_1000\n value: 53.74\n - type: ndcg_at_3\n value: 43.261\n - type: ndcg_at_5\n value: 44.891\n - type: precision_at_1\n value: 39.427\n - type: precision_at_10\n value: 8.847\n - type: precision_at_100\n value: 1.425\n - type: precision_at_1000\n value: 0.189\n - type: precision_at_3\n value: 20.785999999999998\n - type: precision_at_5\n value: 14.560999999999998\n - type: recall_at_1\n value: 31.169999999999998\n - type: recall_at_10\n value: 56.971000000000004\n - type: recall_at_100\n value: 76.31400000000001\n - type: recall_at_1000\n value: 88.93900000000001\n - type: recall_at_3\n value: 45.208\n - type: recall_at_5\n value: 49.923\n - type: map_at_1\n value: 39.682\n - type: map_at_10\n value: 52.766000000000005\n - type: map_at_100\n value: 53.84100000000001\n - type: map_at_1000\n value: 53.898\n - type: map_at_3\n value: 49.291000000000004\n - type: map_at_5\n value: 51.365\n - type: mrr_at_1\n value: 45.266\n - type: mrr_at_10\n value: 56.093\n - type: mrr_at_100\n value: 56.763\n - type: mrr_at_1000\n value: 56.793000000000006\n - type: mrr_at_3\n value: 53.668000000000006\n - type: mrr_at_5\n value: 55.1\n - type: ndcg_at_1\n value: 45.266\n - type: ndcg_at_10\n value: 58.836\n - type: ndcg_at_100\n value: 62.863\n - type: ndcg_at_1000\n value: 63.912\n - type: ndcg_at_3\n value: 53.19199999999999\n - type: ndcg_at_5\n value: 56.125\n - type: precision_at_1\n value: 45.266\n - type: precision_at_10\n value: 9.492\n - type: precision_at_100\n value: 1.236\n - type: precision_at_1000\n value: 0.13699999999999998\n - type: precision_at_3\n value: 23.762\n - type: precision_at_5\n value: 16.414\n - type: recall_at_1\n value: 39.682\n - type: recall_at_10\n value: 73.233\n - type: recall_at_100\n value: 90.335\n - type: recall_at_1000\n value: 97.452\n - type: recall_at_3\n value: 58.562000000000005\n - type: recall_at_5\n value: 65.569\n - type: map_at_1\n value: 26.743\n - type: map_at_10\n value: 34.016000000000005\n - type: map_at_100\n value: 35.028999999999996\n - type: map_at_1000\n value: 35.113\n - type: map_at_3\n value: 31.763\n - type: map_at_5\n value: 33.013999999999996\n - type: mrr_at_1\n value: 28.927000000000003\n - type: mrr_at_10\n value: 36.32\n - type: mrr_at_100\n value: 37.221\n - type: mrr_at_1000\n value: 37.281\n - type: mrr_at_3\n value: 34.105000000000004\n - type: mrr_at_5\n value: 35.371\n - type: ndcg_at_1\n value: 28.927000000000003\n - type: ndcg_at_10\n value: 38.474000000000004\n - type: ndcg_at_100\n value: 43.580000000000005\n - type: ndcg_at_1000\n value: 45.64\n - type: ndcg_at_3\n value: 34.035\n - type: ndcg_at_5\n value: 36.186\n - type: precision_at_1\n value: 28.927000000000003\n - type: precision_at_10\n value: 5.74\n - type: precision_at_100\n value: 0.8710000000000001\n - type: precision_at_1000\n value: 0.108\n - type: precision_at_3\n value: 14.124\n - type: precision_at_5\n value: 9.74\n - type: recall_at_1\n value: 26.743\n - type: recall_at_10\n value: 49.955\n - type: recall_at_100\n value: 73.904\n - type: recall_at_1000\n value: 89.133\n - type: recall_at_3\n value: 38.072\n - type: recall_at_5\n value: 43.266\n - type: map_at_1\n value: 16.928\n - type: map_at_10\n value: 23.549\n - type: map_at_100\n value: 24.887\n - type: map_at_1000\n value: 25.018\n - type: map_at_3\n value: 21.002000000000002\n - type: map_at_5\n value: 22.256\n - type: mrr_at_1\n value: 21.02\n - type: mrr_at_10\n value: 27.898\n - type: mrr_at_100\n value: 29.018\n - type: mrr_at_1000\n value: 29.099999999999998\n - type: mrr_at_3\n value: 25.456\n - type: mrr_at_5\n value: 26.625\n - type: ndcg_at_1\n value: 21.02\n - type: ndcg_at_10\n value: 28.277\n - type: ndcg_at_100\n value: 34.54\n - type: ndcg_at_1000\n value: 37.719\n - type: ndcg_at_3\n value: 23.707\n - type: ndcg_at_5\n value: 25.482\n - type: precision_at_1\n value: 21.02\n - type: precision_at_10\n value: 5.361\n - type: precision_at_100\n value: 0.9809999999999999\n - type: precision_at_1000\n value: 0.13899999999999998\n - type: precision_at_3\n value: 11.401\n - type: precision_at_5\n value: 8.209\n - type: recall_at_1\n value: 16.928\n - type: recall_at_10\n value: 38.601\n - type: recall_at_100\n value: 65.759\n - type: recall_at_1000\n value: 88.543\n - type: recall_at_3\n value: 25.556\n - type: recall_at_5\n value: 30.447000000000003\n - type: map_at_1\n value: 28.549000000000003\n - type: map_at_10\n value: 38.426\n - type: map_at_100\n value: 39.845000000000006\n - type: map_at_1000\n value: 39.956\n - type: map_at_3\n value: 35.372\n - type: map_at_5\n value: 37.204\n - type: mrr_at_1\n value: 35.034\n - type: mrr_at_10\n value: 44.041000000000004\n - type: mrr_at_100\n value: 44.95\n - type: mrr_at_1000\n value: 44.997\n - type: mrr_at_3\n value: 41.498000000000005\n - type: mrr_at_5\n value: 43.077\n - type: ndcg_at_1\n value: 35.034\n - type: ndcg_at_10\n value: 44.218\n - type: ndcg_at_100\n value: 49.958000000000006\n - type: ndcg_at_1000\n value: 52.019000000000005\n - type: ndcg_at_3\n value: 39.34\n - type: ndcg_at_5\n value: 41.892\n - type: precision_at_1\n value: 35.034\n - type: precision_at_10\n value: 7.911\n - type: precision_at_100\n value: 1.26\n - type: precision_at_1000\n value: 0.16\n - type: precision_at_3\n value: 18.511\n - type: precision_at_5\n value: 13.205\n - type: recall_at_1\n value: 28.549000000000003\n - type: recall_at_10\n value: 56.035999999999994\n - type: recall_at_100\n value: 79.701\n - type: recall_at_1000\n value: 93.149\n - type: recall_at_3\n value: 42.275\n - type: recall_at_5\n value: 49.097\n - type: map_at_1\n value: 29.391000000000002\n - type: map_at_10\n value: 39.48\n - type: map_at_100\n value: 40.727000000000004\n - type: map_at_1000\n value: 40.835\n - type: map_at_3\n value: 36.234\n - type: map_at_5\n value: 37.877\n - type: mrr_at_1\n value: 35.959\n - type: mrr_at_10\n value: 44.726\n - type: mrr_at_100\n value: 45.531\n - type: mrr_at_1000\n value: 45.582\n - type: mrr_at_3\n value: 42.047000000000004\n - type: mrr_at_5\n value: 43.611\n - type: ndcg_at_1\n value: 35.959\n - type: ndcg_at_10\n value: 45.303\n - type: ndcg_at_100\n value: 50.683\n - type: ndcg_at_1000\n value: 52.818\n - type: ndcg_at_3\n value: 39.987\n - type: ndcg_at_5\n value: 42.243\n - type: precision_at_1\n value: 35.959\n - type: precision_at_10\n value: 8.241999999999999\n - type: precision_at_100\n value: 1.274\n - type: precision_at_1000\n value: 0.163\n - type: precision_at_3\n value: 18.836\n - type: precision_at_5\n value: 13.196\n - type: recall_at_1\n value: 29.391000000000002\n - type: recall_at_10\n value: 57.364000000000004\n - type: recall_at_100\n value: 80.683\n - type: recall_at_1000\n value: 94.918\n - type: recall_at_3\n value: 42.263\n - type: recall_at_5\n value: 48.634\n - type: map_at_1\n value: 26.791749999999997\n - type: map_at_10\n value: 35.75541666666667\n - type: map_at_100\n value: 37.00791666666667\n - type: map_at_1000\n value: 37.12408333333333\n - type: map_at_3\n value: 33.02966666666667\n - type: map_at_5\n value: 34.56866666666667\n - type: mrr_at_1\n value: 31.744333333333337\n - type: mrr_at_10\n value: 39.9925\n - type: mrr_at_100\n value: 40.86458333333333\n - type: mrr_at_1000\n value: 40.92175000000001\n - type: mrr_at_3\n value: 37.68183333333334\n - type: mrr_at_5\n value: 39.028499999999994\n - type: ndcg_at_1\n value: 31.744333333333337\n - type: ndcg_at_10\n value: 40.95008333333334\n - type: ndcg_at_100\n value: 46.25966666666667\n - type: ndcg_at_1000\n value: 48.535333333333334\n - type: ndcg_at_3\n value: 36.43333333333333\n - type: ndcg_at_5\n value: 38.602333333333334\n - type: precision_at_1\n value: 31.744333333333337\n - type: precision_at_10\n value: 7.135166666666666\n - type: precision_at_100\n value: 1.1535833333333334\n - type: precision_at_1000\n value: 0.15391666666666665\n - type: precision_at_3\n value: 16.713\n - type: precision_at_5\n value: 11.828416666666666\n - type: recall_at_1\n value: 26.791749999999997\n - type: recall_at_10\n value: 51.98625\n - type: recall_at_100\n value: 75.30358333333334\n - type: recall_at_1000\n value: 91.05433333333333\n - type: recall_at_3\n value: 39.39583333333333\n - type: recall_at_5\n value: 45.05925\n - type: map_at_1\n value: 22.219\n - type: map_at_10\n value: 29.162\n - type: map_at_100\n value: 30.049999999999997\n - type: map_at_1000\n value: 30.144\n - type: map_at_3\n value: 27.204\n - type: map_at_5\n value: 28.351\n - type: mrr_at_1\n value: 25.153\n - type: mrr_at_10\n value: 31.814999999999998\n - type: mrr_at_100\n value: 32.573\n - type: mrr_at_1000\n value: 32.645\n - type: mrr_at_3\n value: 29.934\n - type: mrr_at_5\n value: 30.946\n - type: ndcg_at_1\n value: 25.153\n - type: ndcg_at_10\n value: 33.099000000000004\n - type: ndcg_at_100\n value: 37.768\n - type: ndcg_at_1000\n value: 40.331\n - type: ndcg_at_3\n value: 29.473\n - type: ndcg_at_5\n value: 31.206\n - type: precision_at_1\n value: 25.153\n - type: precision_at_10\n value: 5.183999999999999\n - type: precision_at_100\n value: 0.8170000000000001\n - type: precision_at_1000\n value: 0.11100000000000002\n - type: precision_at_3\n value: 12.831999999999999\n - type: precision_at_5\n value: 8.895999999999999\n - type: recall_at_1\n value: 22.219\n - type: recall_at_10\n value: 42.637\n - type: recall_at_100\n value: 64.704\n - type: recall_at_1000\n value: 83.963\n - type: recall_at_3\n value: 32.444\n - type: recall_at_5\n value: 36.802\n - type: map_at_1\n value: 17.427999999999997\n - type: map_at_10\n value: 24.029\n - type: map_at_100\n value: 25.119999999999997\n - type: map_at_1000\n value: 25.257\n - type: map_at_3\n value: 22.016\n - type: map_at_5\n value: 23.143\n - type: mrr_at_1\n value: 21.129\n - type: mrr_at_10\n value: 27.750000000000004\n - type: mrr_at_100\n value: 28.666999999999998\n - type: mrr_at_1000\n value: 28.754999999999995\n - type: mrr_at_3\n value: 25.849\n - type: mrr_at_5\n value: 26.939999999999998\n - type: ndcg_at_1\n value: 21.129\n - type: ndcg_at_10\n value: 28.203\n - type: ndcg_at_100\n value: 33.44\n - type: ndcg_at_1000\n value: 36.61\n - type: ndcg_at_3\n value: 24.648999999999997\n - type: ndcg_at_5\n value: 26.316\n - type: precision_at_1\n value: 21.129\n - type: precision_at_10\n value: 5.055\n - type: precision_at_100\n value: 0.909\n - type: precision_at_1000\n value: 0.13699999999999998\n - type: precision_at_3\n value: 11.666\n - type: precision_at_5\n value: 8.3\n - type: recall_at_1\n value: 17.427999999999997\n - type: recall_at_10\n value: 36.923\n - type: recall_at_100\n value: 60.606\n - type: recall_at_1000\n value: 83.19\n - type: recall_at_3\n value: 26.845000000000002\n - type: recall_at_5\n value: 31.247000000000003\n - type: map_at_1\n value: 26.457000000000004\n - type: map_at_10\n value: 35.228\n - type: map_at_100\n value: 36.475\n - type: map_at_1000\n value: 36.585\n - type: map_at_3\n value: 32.444\n - type: map_at_5\n value: 34.046\n - type: mrr_at_1\n value: 30.784\n - type: mrr_at_10\n value: 39.133\n - type: mrr_at_100\n value: 40.11\n - type: mrr_at_1000\n value: 40.169\n - type: mrr_at_3\n value: 36.692\n - type: mrr_at_5\n value: 38.17\n - type: ndcg_at_1\n value: 30.784\n - type: ndcg_at_10\n value: 40.358\n - type: ndcg_at_100\n value: 46.119\n - type: ndcg_at_1000\n value: 48.428\n - type: ndcg_at_3\n value: 35.504000000000005\n - type: ndcg_at_5\n value: 37.864\n - type: precision_at_1\n value: 30.784\n - type: precision_at_10\n value: 6.800000000000001\n - type: precision_at_100\n value: 1.083\n - type: precision_at_1000\n value: 0.13899999999999998\n - type: precision_at_3\n value: 15.920000000000002\n - type: precision_at_5\n value: 11.437\n - type: recall_at_1\n value: 26.457000000000004\n - type: recall_at_10\n value: 51.845\n - type: recall_at_100\n value: 77.046\n - type: recall_at_1000\n value: 92.892\n - type: recall_at_3\n value: 38.89\n - type: recall_at_5\n value: 44.688\n - type: map_at_1\n value: 29.378999999999998\n - type: map_at_10\n value: 37.373\n - type: map_at_100\n value: 39.107\n - type: map_at_1000\n value: 39.317\n - type: map_at_3\n value: 34.563\n - type: map_at_5\n value: 36.173\n - type: mrr_at_1\n value: 35.178\n - type: mrr_at_10\n value: 42.44\n - type: mrr_at_100\n value: 43.434\n - type: mrr_at_1000\n value: 43.482\n - type: mrr_at_3\n value: 39.987\n - type: mrr_at_5\n value: 41.370000000000005\n - type: ndcg_at_1\n value: 35.178\n - type: ndcg_at_10\n value: 42.82\n - type: ndcg_at_100\n value: 48.935\n - type: ndcg_at_1000\n value: 51.28\n - type: ndcg_at_3\n value: 38.562999999999995\n - type: ndcg_at_5\n value: 40.687\n - type: precision_at_1\n value: 35.178\n - type: precision_at_10\n value: 7.945\n - type: precision_at_100\n value: 1.524\n - type: precision_at_1000\n value: 0.242\n - type: precision_at_3\n value: 17.721\n - type: precision_at_5\n value: 12.925\n - type: recall_at_1\n value: 29.378999999999998\n - type: recall_at_10\n value: 52.141999999999996\n - type: recall_at_100\n value: 79.49000000000001\n - type: recall_at_1000\n value: 93.782\n - type: recall_at_3\n value: 39.579\n - type: recall_at_5\n value: 45.462\n - type: map_at_1\n value: 19.814999999999998\n - type: map_at_10\n value: 27.383999999999997\n - type: map_at_100\n value: 28.483999999999998\n - type: map_at_1000\n value: 28.585\n - type: map_at_3\n value: 24.807000000000002\n - type: map_at_5\n value: 26.485999999999997\n - type: mrr_at_1\n value: 21.996\n - type: mrr_at_10\n value: 29.584\n - type: mrr_at_100\n value: 30.611\n - type: mrr_at_1000\n value: 30.684\n - type: mrr_at_3\n value: 27.11\n - type: mrr_at_5\n value: 28.746\n - type: ndcg_at_1\n value: 21.996\n - type: ndcg_at_10\n value: 32.024\n - type: ndcg_at_100\n value: 37.528\n - type: ndcg_at_1000\n value: 40.150999999999996\n - type: ndcg_at_3\n value: 27.016000000000002\n - type: ndcg_at_5\n value: 29.927999999999997\n - type: precision_at_1\n value: 21.996\n - type: precision_at_10\n value: 5.102\n - type: precision_at_100\n value: 0.856\n - type: precision_at_1000\n value: 0.117\n - type: precision_at_3\n value: 11.583\n - type: precision_at_5\n value: 8.577\n - type: recall_at_1\n value: 19.814999999999998\n - type: recall_at_10\n value: 44.239\n - type: recall_at_100\n value: 69.269\n - type: recall_at_1000\n value: 89.216\n - type: recall_at_3\n value: 31.102999999999998\n - type: recall_at_5\n value: 38.078\n - task:\n type: Retrieval\n dataset:\n name: MTEB ClimateFEVER\n type: climate-fever\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 11.349\n - type: map_at_10\n value: 19.436\n - type: map_at_100\n value: 21.282999999999998\n - type: map_at_1000\n value: 21.479\n - type: map_at_3\n value: 15.841\n - type: map_at_5\n value: 17.558\n - type: mrr_at_1\n value: 25.863000000000003\n - type: mrr_at_10\n value: 37.218\n - type: mrr_at_100\n value: 38.198\n - type: mrr_at_1000\n value: 38.236\n - type: mrr_at_3\n value: 33.409\n - type: mrr_at_5\n value: 35.602000000000004\n - type: ndcg_at_1\n value: 25.863000000000003\n - type: ndcg_at_10\n value: 27.953\n - type: ndcg_at_100\n value: 35.327\n - type: ndcg_at_1000\n value: 38.708999999999996\n - type: ndcg_at_3\n value: 21.985\n - type: ndcg_at_5\n value: 23.957\n - type: precision_at_1\n value: 25.863000000000003\n - type: precision_at_10\n value: 8.99\n - type: precision_at_100\n value: 1.6889999999999998\n - type: precision_at_1000\n value: 0.232\n - type: precision_at_3\n value: 16.308\n - type: precision_at_5\n value: 12.912\n - type: recall_at_1\n value: 11.349\n - type: recall_at_10\n value: 34.581\n - type: recall_at_100\n value: 60.178\n - type: recall_at_1000\n value: 78.88199999999999\n - type: recall_at_3\n value: 20.041999999999998\n - type: recall_at_5\n value: 25.458\n - task:\n type: Retrieval\n dataset:\n name: MTEB DBPedia\n type: dbpedia-entity\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 7.893\n - type: map_at_10\n value: 15.457\n - type: map_at_100\n value: 20.905\n - type: map_at_1000\n value: 22.116\n - type: map_at_3\n value: 11.593\n - type: map_at_5\n value: 13.134\n - type: mrr_at_1\n value: 57.49999999999999\n - type: mrr_at_10\n value: 65.467\n - type: mrr_at_100\n value: 66.022\n - type: mrr_at_1000\n value: 66.039\n - type: mrr_at_3\n value: 63.458000000000006\n - type: mrr_at_5\n value: 64.546\n - type: ndcg_at_1\n value: 45.875\n - type: ndcg_at_10\n value: 33.344\n - type: ndcg_at_100\n value: 36.849\n - type: ndcg_at_1000\n value: 44.03\n - type: ndcg_at_3\n value: 37.504\n - type: ndcg_at_5\n value: 34.892\n - type: precision_at_1\n value: 57.49999999999999\n - type: precision_at_10\n value: 25.95\n - type: precision_at_100\n value: 7.89\n - type: precision_at_1000\n value: 1.669\n - type: precision_at_3\n value: 40.333000000000006\n - type: precision_at_5\n value: 33.050000000000004\n - type: recall_at_1\n value: 7.893\n - type: recall_at_10\n value: 20.724999999999998\n - type: recall_at_100\n value: 42.516\n - type: recall_at_1000\n value: 65.822\n - type: recall_at_3\n value: 12.615000000000002\n - type: recall_at_5\n value: 15.482000000000001\n - task:\n type: Classification\n dataset:\n name: MTEB EmotionClassification\n type: mteb/emotion\n config: default\n split: test\n revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37\n metrics:\n - type: accuracy\n value: 51.760000000000005\n - type: f1\n value: 45.51690565701713\n - task:\n type: Retrieval\n dataset:\n name: MTEB FEVER\n type: fever\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 53.882\n - type: map_at_10\n value: 65.902\n - type: map_at_100\n value: 66.33\n - type: map_at_1000\n value: 66.348\n - type: map_at_3\n value: 63.75999999999999\n - type: map_at_5\n value: 65.181\n - type: mrr_at_1\n value: 58.041\n - type: mrr_at_10\n value: 70.133\n - type: mrr_at_100\n value: 70.463\n - type: mrr_at_1000\n value: 70.47\n - type: mrr_at_3\n value: 68.164\n - type: mrr_at_5\n value: 69.465\n - type: ndcg_at_1\n value: 58.041\n - type: ndcg_at_10\n value: 71.84700000000001\n - type: ndcg_at_100\n value: 73.699\n - type: ndcg_at_1000\n value: 74.06700000000001\n - type: ndcg_at_3\n value: 67.855\n - type: ndcg_at_5\n value: 70.203\n - type: precision_at_1\n value: 58.041\n - type: precision_at_10\n value: 9.427000000000001\n - type: precision_at_100\n value: 1.049\n - type: precision_at_1000\n value: 0.11\n - type: precision_at_3\n value: 27.278000000000002\n - type: precision_at_5\n value: 17.693\n - type: recall_at_1\n value: 53.882\n - type: recall_at_10\n value: 85.99\n - type: recall_at_100\n value: 94.09100000000001\n - type: recall_at_1000\n value: 96.612\n - type: recall_at_3\n value: 75.25\n - type: recall_at_5\n value: 80.997\n - task:\n type: Retrieval\n dataset:\n name: MTEB FiQA2018\n type: fiqa\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 19.165\n - type: map_at_10\n value: 31.845000000000002\n - type: map_at_100\n value: 33.678999999999995\n - type: map_at_1000\n value: 33.878\n - type: map_at_3\n value: 27.881\n - type: map_at_5\n value: 30.049999999999997\n - type: mrr_at_1\n value: 38.272\n - type: mrr_at_10\n value: 47.04\n - type: mrr_at_100\n value: 47.923\n - type: mrr_at_1000\n value: 47.973\n - type: mrr_at_3\n value: 44.985\n - type: mrr_at_5\n value: 46.150000000000006\n - type: ndcg_at_1\n value: 38.272\n - type: ndcg_at_10\n value: 39.177\n - type: ndcg_at_100\n value: 45.995000000000005\n - type: ndcg_at_1000\n value: 49.312\n - type: ndcg_at_3\n value: 36.135\n - type: ndcg_at_5\n value: 36.936\n - type: precision_at_1\n value: 38.272\n - type: precision_at_10\n value: 10.926\n - type: precision_at_100\n value: 1.809\n - type: precision_at_1000\n value: 0.23700000000000002\n - type: precision_at_3\n value: 24.331\n - type: precision_at_5\n value: 17.747\n - type: recall_at_1\n value: 19.165\n - type: recall_at_10\n value: 45.103\n - type: recall_at_100\n value: 70.295\n - type: recall_at_1000\n value: 90.592\n - type: recall_at_3\n value: 32.832\n - type: recall_at_5\n value: 37.905\n - task:\n type: Retrieval\n dataset:\n name: MTEB HotpotQA\n type: hotpotqa\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 32.397\n - type: map_at_10\n value: 44.83\n - type: map_at_100\n value: 45.716\n - type: map_at_1000\n value: 45.797\n - type: map_at_3\n value: 41.955999999999996\n - type: map_at_5\n value: 43.736999999999995\n - type: mrr_at_1\n value: 64.794\n - type: mrr_at_10\n value: 71.866\n - type: mrr_at_100\n value: 72.22\n - type: mrr_at_1000\n value: 72.238\n - type: mrr_at_3\n value: 70.416\n - type: mrr_at_5\n value: 71.304\n - type: ndcg_at_1\n value: 64.794\n - type: ndcg_at_10\n value: 54.186\n - type: ndcg_at_100\n value: 57.623000000000005\n - type: ndcg_at_1000\n value: 59.302\n - type: ndcg_at_3\n value: 49.703\n - type: ndcg_at_5\n value: 52.154999999999994\n - type: precision_at_1\n value: 64.794\n - type: precision_at_10\n value: 11.219\n - type: precision_at_100\n value: 1.394\n - type: precision_at_1000\n value: 0.16199999999999998\n - type: precision_at_3\n value: 30.767\n - type: precision_at_5\n value: 20.397000000000002\n - type: recall_at_1\n value: 32.397\n - type: recall_at_10\n value: 56.096999999999994\n - type: recall_at_100\n value: 69.696\n - type: recall_at_1000\n value: 80.88499999999999\n - type: recall_at_3\n value: 46.150999999999996\n - type: recall_at_5\n value: 50.993\n - task:\n type: Classification\n dataset:\n name: MTEB ImdbClassification\n type: mteb/imdb\n config: default\n split: test\n revision: 3d86128a09e091d6018b6d26cad27f2739fc2db7\n metrics:\n - type: accuracy\n value: 81.1744\n - type: ap\n value: 75.44973697032414\n - type: f1\n value: 81.09901117955782\n - task:\n type: Retrieval\n dataset:\n name: MTEB MSMARCO\n type: msmarco\n config: default\n split: dev\n revision: None\n metrics:\n - type: map_at_1\n value: 19.519000000000002\n - type: map_at_10\n value: 31.025000000000002\n - type: map_at_100\n value: 32.275999999999996\n - type: map_at_1000\n value: 32.329\n - type: map_at_3\n value: 27.132\n - type: map_at_5\n value: 29.415999999999997\n - type: mrr_at_1\n value: 20.115\n - type: mrr_at_10\n value: 31.569000000000003\n - type: mrr_at_100\n value: 32.768\n - type: mrr_at_1000\n value: 32.816\n - type: mrr_at_3\n value: 27.748\n - type: mrr_at_5\n value: 29.956\n - type: ndcg_at_1\n value: 20.115\n - type: ndcg_at_10\n value: 37.756\n - type: ndcg_at_100\n value: 43.858000000000004\n - type: ndcg_at_1000\n value: 45.199\n - type: ndcg_at_3\n value: 29.818\n - type: ndcg_at_5\n value: 33.875\n - type: precision_at_1\n value: 20.115\n - type: precision_at_10\n value: 6.122\n - type: precision_at_100\n value: 0.919\n - type: precision_at_1000\n value: 0.10300000000000001\n - type: precision_at_3\n value: 12.794\n - type: precision_at_5\n value: 9.731\n - type: recall_at_1\n value: 19.519000000000002\n - type: recall_at_10\n value: 58.62500000000001\n - type: recall_at_100\n value: 86.99\n - type: recall_at_1000\n value: 97.268\n - type: recall_at_3\n value: 37.002\n - type: recall_at_5\n value: 46.778\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPDomainClassification (en)\n type: mteb/mtop_domain\n config: en\n split: test\n revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf\n metrics:\n - type: accuracy\n value: 93.71865025079799\n - type: f1\n value: 93.38906173610519\n - task:\n type: Classification\n dataset:\n name: MTEB MTOPIntentClassification (en)\n type: mteb/mtop_intent\n config: en\n split: test\n revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba\n metrics:\n - type: accuracy\n value: 70.2576379388965\n - type: f1\n value: 49.20405830249464\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveIntentClassification (en)\n type: mteb/amazon_massive_intent\n config: en\n split: test\n revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7\n metrics:\n - type: accuracy\n value: 67.48486886348351\n - type: f1\n value: 64.92199176095157\n - task:\n type: Classification\n dataset:\n name: MTEB MassiveScenarioClassification (en)\n type: mteb/amazon_massive_scenario\n config: en\n split: test\n revision: 7d571f92784cd94a019292a1f45445077d0ef634\n metrics:\n - type: accuracy\n value: 72.59246805648958\n - type: f1\n value: 72.1222026389164\n - task:\n type: Clustering\n dataset:\n name: MTEB MedrxivClusteringP2P\n type: mteb/medrxiv-clustering-p2p\n config: default\n split: test\n revision: e7a26af6f3ae46b30dde8737f02c07b1505bcc73\n metrics:\n - type: v_measure\n value: 30.887642595096825\n - task:\n type: Clustering\n dataset:\n name: MTEB MedrxivClusteringS2S\n type: mteb/medrxiv-clustering-s2s\n config: default\n split: test\n revision: 35191c8c0dca72d8ff3efcd72aa802307d469663\n metrics:\n - type: v_measure\n value: 28.3764418784054\n - task:\n type: Reranking\n dataset:\n name: MTEB MindSmallReranking\n type: mteb/mind_small\n config: default\n split: test\n revision: 3bdac13927fdc888b903db93b2ffdbd90b295a69\n metrics:\n - type: map\n value: 31.81544126336991\n - type: mrr\n value: 32.82666576268031\n - task:\n type: Retrieval\n dataset:\n name: MTEB NFCorpus\n type: nfcorpus\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 5.185\n - type: map_at_10\n value: 11.158\n - type: map_at_100\n value: 14.041\n - type: map_at_1000\n value: 15.360999999999999\n - type: map_at_3\n value: 8.417\n - type: map_at_5\n value: 9.378\n - type: mrr_at_1\n value: 44.582\n - type: mrr_at_10\n value: 53.083999999999996\n - type: mrr_at_100\n value: 53.787\n - type: mrr_at_1000\n value: 53.824000000000005\n - type: mrr_at_3\n value: 51.187000000000005\n - type: mrr_at_5\n value: 52.379\n - type: ndcg_at_1\n value: 42.57\n - type: ndcg_at_10\n value: 31.593\n - type: ndcg_at_100\n value: 29.093999999999998\n - type: ndcg_at_1000\n value: 37.909\n - type: ndcg_at_3\n value: 37.083\n - type: ndcg_at_5\n value: 34.397\n - type: precision_at_1\n value: 43.963\n - type: precision_at_10\n value: 23.498\n - type: precision_at_100\n value: 7.6160000000000005\n - type: precision_at_1000\n value: 2.032\n - type: precision_at_3\n value: 34.572\n - type: precision_at_5\n value: 29.412\n - type: recall_at_1\n value: 5.185\n - type: recall_at_10\n value: 15.234\n - type: recall_at_100\n value: 29.49\n - type: recall_at_1000\n value: 62.273999999999994\n - type: recall_at_3\n value: 9.55\n - type: recall_at_5\n value: 11.103\n - task:\n type: Retrieval\n dataset:\n name: MTEB NQ\n type: nq\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 23.803\n - type: map_at_10\n value: 38.183\n - type: map_at_100\n value: 39.421\n - type: map_at_1000\n value: 39.464\n - type: map_at_3\n value: 33.835\n - type: map_at_5\n value: 36.327\n - type: mrr_at_1\n value: 26.68\n - type: mrr_at_10\n value: 40.439\n - type: mrr_at_100\n value: 41.415\n - type: mrr_at_1000\n value: 41.443999999999996\n - type: mrr_at_3\n value: 36.612\n - type: mrr_at_5\n value: 38.877\n - type: ndcg_at_1\n value: 26.68\n - type: ndcg_at_10\n value: 45.882\n - type: ndcg_at_100\n value: 51.227999999999994\n - type: ndcg_at_1000\n value: 52.207\n - type: ndcg_at_3\n value: 37.511\n - type: ndcg_at_5\n value: 41.749\n - type: precision_at_1\n value: 26.68\n - type: precision_at_10\n value: 7.9750000000000005\n - type: precision_at_100\n value: 1.0959999999999999\n - type: precision_at_1000\n value: 0.11900000000000001\n - type: precision_at_3\n value: 17.449\n - type: precision_at_5\n value: 12.897\n - type: recall_at_1\n value: 23.803\n - type: recall_at_10\n value: 67.152\n - type: recall_at_100\n value: 90.522\n - type: recall_at_1000\n value: 97.743\n - type: recall_at_3\n value: 45.338\n - type: recall_at_5\n value: 55.106\n - task:\n type: Retrieval\n dataset:\n name: MTEB QuoraRetrieval\n type: quora\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 70.473\n - type: map_at_10\n value: 84.452\n - type: map_at_100\n value: 85.101\n - type: map_at_1000\n value: 85.115\n - type: map_at_3\n value: 81.435\n - type: map_at_5\n value: 83.338\n - type: mrr_at_1\n value: 81.19\n - type: mrr_at_10\n value: 87.324\n - type: mrr_at_100\n value: 87.434\n - type: mrr_at_1000\n value: 87.435\n - type: mrr_at_3\n value: 86.31\n - type: mrr_at_5\n value: 87.002\n - type: ndcg_at_1\n value: 81.21000000000001\n - type: ndcg_at_10\n value: 88.19\n - type: ndcg_at_100\n value: 89.44\n - type: ndcg_at_1000\n value: 89.526\n - type: ndcg_at_3\n value: 85.237\n - type: ndcg_at_5\n value: 86.892\n - type: precision_at_1\n value: 81.21000000000001\n - type: precision_at_10\n value: 13.417000000000002\n - type: precision_at_100\n value: 1.537\n - type: precision_at_1000\n value: 0.157\n - type: precision_at_3\n value: 37.31\n - type: precision_at_5\n value: 24.59\n - type: recall_at_1\n value: 70.473\n - type: recall_at_10\n value: 95.367\n - type: recall_at_100\n value: 99.616\n - type: recall_at_1000\n value: 99.996\n - type: recall_at_3\n value: 86.936\n - type: recall_at_5\n value: 91.557\n - task:\n type: Clustering\n dataset:\n name: MTEB RedditClustering\n type: mteb/reddit-clustering\n config: default\n split: test\n revision: 24640382cdbf8abc73003fb0fa6d111a705499eb\n metrics:\n - type: v_measure\n value: 59.25776525253911\n - task:\n type: Clustering\n dataset:\n name: MTEB RedditClusteringP2P\n type: mteb/reddit-clustering-p2p\n config: default\n split: test\n revision: 282350215ef01743dc01b456c7f5241fa8937f16\n metrics:\n - type: v_measure\n value: 63.22135271663078\n - task:\n type: Retrieval\n dataset:\n name: MTEB SCIDOCS\n type: scidocs\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 4.003\n - type: map_at_10\n value: 10.062999999999999\n - type: map_at_100\n value: 11.854000000000001\n - type: map_at_1000\n value: 12.145999999999999\n - type: map_at_3\n value: 7.242\n - type: map_at_5\n value: 8.652999999999999\n - type: mrr_at_1\n value: 19.7\n - type: mrr_at_10\n value: 29.721999999999998\n - type: mrr_at_100\n value: 30.867\n - type: mrr_at_1000\n value: 30.944\n - type: mrr_at_3\n value: 26.683\n - type: mrr_at_5\n value: 28.498\n - type: ndcg_at_1\n value: 19.7\n - type: ndcg_at_10\n value: 17.095\n - type: ndcg_at_100\n value: 24.375\n - type: ndcg_at_1000\n value: 29.831000000000003\n - type: ndcg_at_3\n value: 16.305\n - type: ndcg_at_5\n value: 14.291\n - type: precision_at_1\n value: 19.7\n - type: precision_at_10\n value: 8.799999999999999\n - type: precision_at_100\n value: 1.9349999999999998\n - type: precision_at_1000\n value: 0.32399999999999995\n - type: precision_at_3\n value: 15.2\n - type: precision_at_5\n value: 12.540000000000001\n - type: recall_at_1\n value: 4.003\n - type: recall_at_10\n value: 17.877000000000002\n - type: recall_at_100\n value: 39.217\n - type: recall_at_1000\n value: 65.862\n - type: recall_at_3\n value: 9.242\n - type: recall_at_5\n value: 12.715000000000002\n - task:\n type: STS\n dataset:\n name: MTEB SICK-R\n type: mteb/sickr-sts\n config: default\n split: test\n revision: a6ea5a8cab320b040a23452cc28066d9beae2cee\n metrics:\n - type: cos_sim_spearman\n value: 80.25888668589654\n - task:\n type: STS\n dataset:\n name: MTEB STS12\n type: mteb/sts12-sts\n config: default\n split: test\n revision: a0d554a64d88156834ff5ae9920b964011b16384\n metrics:\n - type: cos_sim_spearman\n value: 77.02037527837669\n - task:\n type: STS\n dataset:\n name: MTEB STS13\n type: mteb/sts13-sts\n config: default\n split: test\n revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca\n metrics:\n - type: cos_sim_spearman\n value: 86.58432681008449\n - task:\n type: STS\n dataset:\n name: MTEB STS14\n type: mteb/sts14-sts\n config: default\n split: test\n revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375\n metrics:\n - type: cos_sim_spearman\n value: 81.31697756099051\n - task:\n type: STS\n dataset:\n name: MTEB STS15\n type: mteb/sts15-sts\n config: default\n split: test\n revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3\n metrics:\n - type: cos_sim_spearman\n value: 88.18867599667057\n - task:\n type: STS\n dataset:\n name: MTEB STS16\n type: mteb/sts16-sts\n config: default\n split: test\n revision: 4d8694f8f0e0100860b497b999b3dbed754a0513\n metrics:\n - type: cos_sim_spearman\n value: 84.87853941747623\n - task:\n type: STS\n dataset:\n name: MTEB STS17 (en-en)\n type: mteb/sts17-crosslingual-sts\n config: en-en\n split: test\n revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d\n metrics:\n - type: cos_sim_spearman\n value: 89.46479925383916\n - task:\n type: STS\n dataset:\n name: MTEB STS22 (en)\n type: mteb/sts22-crosslingual-sts\n config: en\n split: test\n revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80\n metrics:\n - type: cos_sim_spearman\n value: 66.45272113649146\n - task:\n type: STS\n dataset:\n name: MTEB STSBenchmark\n type: mteb/stsbenchmark-sts\n config: default\n split: test\n revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831\n metrics:\n - type: cos_sim_spearman\n value: 86.43357313527851\n - task:\n type: Reranking\n dataset:\n name: MTEB SciDocsRR\n type: mteb/scidocs-reranking\n config: default\n split: test\n revision: d3c5e1fc0b855ab6097bf1cda04dd73947d7caab\n metrics:\n - type: map\n value: 78.82761687254882\n - type: mrr\n value: 93.46223674655047\n - task:\n type: Retrieval\n dataset:\n name: MTEB SciFact\n type: scifact\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 44.583\n - type: map_at_10\n value: 52.978\n - type: map_at_100\n value: 53.803\n - type: map_at_1000\n value: 53.839999999999996\n - type: map_at_3\n value: 50.03300000000001\n - type: map_at_5\n value: 51.939\n - type: mrr_at_1\n value: 47.0\n - type: mrr_at_10\n value: 54.730000000000004\n - type: mrr_at_100\n value: 55.31399999999999\n - type: mrr_at_1000\n value: 55.346\n - type: mrr_at_3\n value: 52.0\n - type: mrr_at_5\n value: 53.783\n - type: ndcg_at_1\n value: 47.0\n - type: ndcg_at_10\n value: 57.82899999999999\n - type: ndcg_at_100\n value: 61.49400000000001\n - type: ndcg_at_1000\n value: 62.676\n - type: ndcg_at_3\n value: 52.373000000000005\n - type: ndcg_at_5\n value: 55.481\n - type: precision_at_1\n value: 47.0\n - type: precision_at_10\n value: 7.867\n - type: precision_at_100\n value: 0.997\n - type: precision_at_1000\n value: 0.11\n - type: precision_at_3\n value: 20.556\n - type: precision_at_5\n value: 14.066999999999998\n - type: recall_at_1\n value: 44.583\n - type: recall_at_10\n value: 71.172\n - type: recall_at_100\n value: 87.7\n - type: recall_at_1000\n value: 97.333\n - type: recall_at_3\n value: 56.511\n - type: recall_at_5\n value: 64.206\n - task:\n type: PairClassification\n dataset:\n name: MTEB SprintDuplicateQuestions\n type: mteb/sprintduplicatequestions-pairclassification\n config: default\n split: test\n revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46\n metrics:\n - type: cos_sim_accuracy\n value: 99.66237623762376\n - type: cos_sim_ap\n value: 90.35465126226322\n - type: cos_sim_f1\n value: 82.44575936883628\n - type: cos_sim_precision\n value: 81.32295719844358\n - type: cos_sim_recall\n value: 83.6\n - type: dot_accuracy\n value: 99.66237623762376\n - type: dot_ap\n value: 90.35464287920453\n - type: dot_f1\n value: 82.44575936883628\n - type: dot_precision\n value: 81.32295719844358\n - type: dot_recall\n value: 83.6\n - type: euclidean_accuracy\n value: 99.66237623762376\n - type: euclidean_ap\n value: 90.3546512622632\n - type: euclidean_f1\n value: 82.44575936883628\n - type: euclidean_precision\n value: 81.32295719844358\n - type: euclidean_recall\n value: 83.6\n - type: manhattan_accuracy\n value: 99.65940594059406\n - type: manhattan_ap\n value: 90.29220174849843\n - type: manhattan_f1\n value: 82.4987605354487\n - type: manhattan_precision\n value: 81.80924287118977\n - type: manhattan_recall\n value: 83.2\n - type: max_accuracy\n value: 99.66237623762376\n - type: max_ap\n value: 90.35465126226322\n - type: max_f1\n value: 82.4987605354487\n - task:\n type: Clustering\n dataset:\n name: MTEB StackExchangeClustering\n type: mteb/stackexchange-clustering\n config: default\n split: test\n revision: 6cbc1f7b2bc0622f2e39d2c77fa502909748c259\n metrics:\n - type: v_measure\n value: 65.0394225901397\n - task:\n type: Clustering\n dataset:\n name: MTEB StackExchangeClusteringP2P\n type: mteb/stackexchange-clustering-p2p\n config: default\n split: test\n revision: 815ca46b2622cec33ccafc3735d572c266efdb44\n metrics:\n - type: v_measure\n value: 35.27954189859326\n - task:\n type: Reranking\n dataset:\n name: MTEB StackOverflowDupQuestions\n type: mteb/stackoverflowdupquestions-reranking\n config: default\n split: test\n revision: e185fbe320c72810689fc5848eb6114e1ef5ec69\n metrics:\n - type: map\n value: 50.99055979974896\n - type: mrr\n value: 51.82745257193787\n - task:\n type: Summarization\n dataset:\n name: MTEB SummEval\n type: mteb/summeval\n config: default\n split: test\n revision: cda12ad7615edc362dbf25a00fdd61d3b1eaf93c\n metrics:\n - type: cos_sim_pearson\n value: 30.21655465344237\n - type: cos_sim_spearman\n value: 29.853205339630172\n - type: dot_pearson\n value: 30.216540628083564\n - type: dot_spearman\n value: 29.868978894753027\n - task:\n type: Retrieval\n dataset:\n name: MTEB TRECCOVID\n type: trec-covid\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 0.2\n - type: map_at_10\n value: 1.398\n - type: map_at_100\n value: 7.406\n - type: map_at_1000\n value: 18.401\n - type: map_at_3\n value: 0.479\n - type: map_at_5\n value: 0.772\n - type: mrr_at_1\n value: 70.0\n - type: mrr_at_10\n value: 79.25999999999999\n - type: mrr_at_100\n value: 79.25999999999999\n - type: mrr_at_1000\n value: 79.25999999999999\n - type: mrr_at_3\n value: 77.333\n - type: mrr_at_5\n value: 78.133\n - type: ndcg_at_1\n value: 63.0\n - type: ndcg_at_10\n value: 58.548\n - type: ndcg_at_100\n value: 45.216\n - type: ndcg_at_1000\n value: 41.149\n - type: ndcg_at_3\n value: 60.641999999999996\n - type: ndcg_at_5\n value: 61.135\n - type: precision_at_1\n value: 70.0\n - type: precision_at_10\n value: 64.0\n - type: precision_at_100\n value: 46.92\n - type: precision_at_1000\n value: 18.642\n - type: precision_at_3\n value: 64.667\n - type: precision_at_5\n value: 66.4\n - type: recall_at_1\n value: 0.2\n - type: recall_at_10\n value: 1.6729999999999998\n - type: recall_at_100\n value: 10.856\n - type: recall_at_1000\n value: 38.964999999999996\n - type: recall_at_3\n value: 0.504\n - type: recall_at_5\n value: 0.852\n - task:\n type: Retrieval\n dataset:\n name: MTEB Touche2020\n type: webis-touche2020\n config: default\n split: test\n revision: None\n metrics:\n - type: map_at_1\n value: 1.6629999999999998\n - type: map_at_10\n value: 8.601\n - type: map_at_100\n value: 14.354\n - type: map_at_1000\n value: 15.927\n - type: map_at_3\n value: 4.1930000000000005\n - type: map_at_5\n value: 5.655\n - type: mrr_at_1\n value: 18.367\n - type: mrr_at_10\n value: 34.466\n - type: mrr_at_100\n value: 35.235\n - type: mrr_at_1000\n value: 35.27\n - type: mrr_at_3\n value: 28.571\n - type: mrr_at_5\n value: 31.531\n - type: ndcg_at_1\n value: 14.285999999999998\n - type: ndcg_at_10\n value: 20.374\n - type: ndcg_at_100\n value: 33.532000000000004\n - type: ndcg_at_1000\n value: 45.561\n - type: ndcg_at_3\n value: 18.442\n - type: ndcg_at_5\n value: 18.076\n - type: precision_at_1\n value: 18.367\n - type: precision_at_10\n value: 20.204\n - type: precision_at_100\n value: 7.489999999999999\n - type: precision_at_1000\n value: 1.5630000000000002\n - type: precision_at_3\n value: 21.769\n - type: precision_at_5\n value: 20.408\n - type: recall_at_1\n value: 1.6629999999999998\n - type: recall_at_10\n value: 15.549\n - type: recall_at_100\n value: 47.497\n - type: recall_at_1000\n value: 84.524\n - type: recall_at_3\n value: 5.289\n - type: recall_at_5\n value: 8.035\n - task:\n type: Classification\n dataset:\n name: MTEB ToxicConversationsClassification\n type: mteb/toxic_conversations_50k\n config: default\n split: test\n revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c\n metrics:\n - type: accuracy\n value: 71.8194\n - type: ap\n value: 14.447702451658554\n - type: f1\n value: 55.13659412856185\n - task:\n type: Classification\n dataset:\n name: MTEB TweetSentimentExtractionClassification\n type: mteb/tweet_sentiment_extraction\n config: default\n split: test\n revision: d604517c81ca91fe16a244d1248fc021f9ecee7a\n metrics:\n - type: accuracy\n value: 63.310696095076416\n - type: f1\n value: 63.360434851097814\n - task:\n type: Clustering\n dataset:\n name: MTEB TwentyNewsgroupsClustering\n type: mteb/twentynewsgroups-clustering\n config: default\n split: test\n revision: 6125ec4e24fa026cec8a478383ee943acfbd5449\n metrics:\n - type: v_measure\n value: 51.30677907335145\n - task:\n type: PairClassification\n dataset:\n name: MTEB TwitterSemEval2015\n type: mteb/twittersemeval2015-pairclassification\n config: default\n split: test\n revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1\n metrics:\n - type: cos_sim_accuracy\n value: 86.12386004649221\n - type: cos_sim_ap\n value: 73.99096426215495\n - type: cos_sim_f1\n value: 68.18416968442834\n - type: cos_sim_precision\n value: 66.86960933536275\n - type: cos_sim_recall\n value: 69.55145118733509\n - type: dot_accuracy\n value: 86.12386004649221\n - type: dot_ap\n value: 73.99096813038672\n - type: dot_f1\n value: 68.18416968442834\n - type: dot_precision\n value: 66.86960933536275\n - type: dot_recall\n value: 69.55145118733509\n - type: euclidean_accuracy\n value: 86.12386004649221\n - type: euclidean_ap\n value: 73.99095984980165\n - type: euclidean_f1\n value: 68.18416968442834\n - type: euclidean_precision\n value: 66.86960933536275\n - type: euclidean_recall\n value: 69.55145118733509\n - type: manhattan_accuracy\n value: 86.09405734040651\n - type: manhattan_ap\n value: 73.96825745608601\n - type: manhattan_f1\n value: 68.13888179729383\n - type: manhattan_precision\n value: 65.99901088031652\n - type: manhattan_recall\n value: 70.42216358839049\n - type: max_accuracy\n value: 86.12386004649221\n - type: max_ap\n value: 73.99096813038672\n - type: max_f1\n value: 68.18416968442834\n - task:\n type: PairClassification\n dataset:\n name: MTEB TwitterURLCorpus\n type: mteb/twitterurlcorpus-pairclassification\n config: default\n split: test\n revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf\n metrics:\n - type: cos_sim_accuracy\n value: 88.99367407924865\n - type: cos_sim_ap\n value: 86.19720829843081\n - type: cos_sim_f1\n value: 78.39889075384951\n - type: cos_sim_precision\n value: 74.5110278818144\n - type: cos_sim_recall\n value: 82.71481367416075\n - type: dot_accuracy\n value: 88.99367407924865\n - type: dot_ap\n value: 86.19718471454047\n - type: dot_f1\n value: 78.39889075384951\n - type: dot_precision\n value: 74.5110278818144\n - type: dot_recall\n value: 82.71481367416075\n - type: euclidean_accuracy\n value: 88.99367407924865\n - type: euclidean_ap\n value: 86.1972021422436\n - type: euclidean_f1\n value: 78.39889075384951\n - type: euclidean_precision\n value: 74.5110278818144\n - type: euclidean_recall\n value: 82.71481367416075\n - type: manhattan_accuracy\n value: 88.95680521597392\n - type: manhattan_ap\n value: 86.16659921351506\n - type: manhattan_f1\n value: 78.39125971550081\n - type: manhattan_precision\n value: 74.82502799552073\n - type: manhattan_recall\n value: 82.31444410224823\n - type: max_accuracy\n value: 88.99367407924865\n - type: max_ap\n value: 86.19720829843081\n - type: max_f1\n value: 78.39889075384951\n---\n\n# hkunlp/instructor-base\nWe introduce **Instructor**👨🏫, an instruction-finetuned text embedding model that can generate text embeddings tailored to any task (e.g., classification, retrieval, clustering, text evaluation, etc.) and domains (e.g., science, finance, etc.) ***by simply providing the task instruction, without any finetuning***. Instructor👨 achieves sota on 70 diverse embedding tasks!\nThe model is easy to use with **our customized** `sentence-transformer` library. For more details, check out [our paper](https://arxiv.org/abs/2212.09741) and [project page](https://instructor-embedding.github.io/)! \n\n**************************** **Updates** ****************************\n\n* 01/21: We released a new [checkpoint](https://huggingface.co/hkunlp/instructor-base) trained with hard negatives, which gives better performance.\n* 12/21: We released our [paper](https://arxiv.org/abs/2212.09741), [code](https://github.com/HKUNLP/instructor-embedding), [checkpoint](https://huggingface.co/hkunlp/instructor-base) and [project page](https://instructor-embedding.github.io/)! Check them out!\n\n## Quick start\n<hr />\n\n## Installation\n```bash\npip install InstructorEmbedding\n```\n\n## Compute your customized embeddings\nThen you can use the model like this to calculate domain-specific and task-aware embeddings:\n```python\nfrom InstructorEmbedding import INSTRUCTOR\nmodel = INSTRUCTOR('hkunlp/instructor-base')\nsentence = \"3D ActionSLAM: wearable person tracking in multi-floor environments\"\ninstruction = \"Represent the Science title:\"\nembeddings = model.encode([[instruction,sentence]])\nprint(embeddings)\n```\n\n## Use cases\n<hr />\n\n## Calculate embeddings for your customized texts\nIf you want to calculate customized embeddings for specific sentences, you may follow the unified template to write instructions: \n\n Represent the `domain` `text_type` for `task_objective`:\n* `domain` is optional, and it specifies the domain of the text, e.g., science, finance, medicine, etc.\n* `text_type` is required, and it specifies the encoding unit, e.g., sentence, document, paragraph, etc.\n* `task_objective` is optional, and it specifies the objective of embedding, e.g., retrieve a document, classify the sentence, etc.\n\n## Calculate Sentence similarities\nYou can further use the model to compute similarities between two groups of sentences, with **customized embeddings**.\n```python\nfrom sklearn.metrics.pairwise import cosine_similarity\nsentences_a = [['Represent the Science sentence: ','Parton energy loss in QCD matter'], \n ['Represent the Financial statement: ','The Federal Reserve on Wednesday raised its benchmark interest rate.']]\nsentences_b = [['Represent the Science sentence: ','The Chiral Phase Transition in Dissipative Dynamics'],\n ['Represent the Financial statement: ','The funds rose less than 0.5 per cent on Friday']]\nembeddings_a = model.encode(sentences_a)\nembeddings_b = model.encode(sentences_b)\nsimilarities = cosine_similarity(embeddings_a,embeddings_b)\nprint(similarities)\n```\n\n## Information Retrieval\nYou can also use **customized embeddings** for information retrieval.\n```python\nimport numpy as np\nfrom sklearn.metrics.pairwise import cosine_similarity\nquery = [['Represent the Wikipedia question for retrieving supporting documents: ','where is the food stored in a yam plant']]\ncorpus = [['Represent the Wikipedia document for retrieval: ','Capitalism has been dominant in the Western world since the end of feudalism, but most feel[who?] that the term \"mixed economies\" more precisely describes most contemporary economies, due to their containing both private-owned and state-owned enterprises. In capitalism, prices determine the demand-supply scale. For example, higher demand for certain goods and services lead to higher prices and lower demand for certain goods lead to lower prices.'],\n ['Represent the Wikipedia document for retrieval: ',\"The disparate impact theory is especially controversial under the Fair Housing Act because the Act regulates many activities relating to housing, insurance, and mortgage loans—and some scholars have argued that the theory's use under the Fair Housing Act, combined with extensions of the Community Reinvestment Act, contributed to rise of sub-prime lending and the crash of the U.S. housing market and ensuing global economic recession\"],\n ['Represent the Wikipedia document for retrieval: ','Disparate impact in United States labor law refers to practices in employment, housing, and other areas that adversely affect one group of people of a protected characteristic more than another, even though rules applied by employers or landlords are formally neutral. Although the protected classes vary by statute, most federal civil rights laws protect based on race, color, religion, national origin, and sex as protected traits, and some laws include disability status and other traits as well.']]\nquery_embeddings = model.encode(query)\ncorpus_embeddings = model.encode(corpus)\nsimilarities = cosine_similarity(query_embeddings,corpus_embeddings)\nretrieved_doc_id = np.argmax(similarities)\nprint(retrieved_doc_id)\n```\n\n## Clustering\nUse **customized embeddings** for clustering texts in groups.\n```python\nimport sklearn.cluster\nsentences = [['Represent the Medicine sentence for clustering: ','Dynamical Scalar Degree of Freedom in Horava-Lifshitz Gravity'],\n ['Represent the Medicine sentence for clustering: ','Comparison of Atmospheric Neutrino Flux Calculations at Low Energies'],\n ['Represent the Medicine sentence for clustering: ','Fermion Bags in the Massive Gross-Neveu Model'],\n ['Represent the Medicine sentence for clustering: ',\"QCD corrections to Associated t-tbar-H production at the Tevatron\"],\n ['Represent the Medicine sentence for clustering: ','A New Analysis of the R Measurements: Resonance Parameters of the Higher, Vector States of Charmonium']]\nembeddings = model.encode(sentences)\nclustering_model = sklearn.cluster.MiniBatchKMeans(n_clusters=2)\nclustering_model.fit(embeddings)\ncluster_assignment = clustering_model.labels_\nprint(cluster_assignment)\n```"},"matched_task":{"kind":"list like","value":["SUMMARIZATION"],"string":"[\n \"SUMMARIZATION\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["BIOSSES","SCIFACT"],"string":"[\n \"BIOSSES\",\n \"SCIFACT\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"Non_BioNLP"}}},{"rowIdx":296,"cells":{"id":{"kind":"string","value":"BSC-LT/salamandra-7b-instruct"},"author":{"kind":"string","value":"BSC-LT"},"task_category":{"kind":"string","value":"text-generation"},"tags":{"kind":"list like","value":["transformers","safetensors","llama","text-generation","conversational","bg","ca","code","cs","cy","da","de","el","en","es","et","eu","fi","fr","ga","gl","hr","hu","it","lt","lv","mt","nl","nn","oc","pl","pt","ro","ru","sh","sk","sl","sr","sv","uk","dataset:oscar-corpus/colossal-oscar-1.0","dataset:HuggingFaceFW/fineweb-edu","dataset:joelniklaus/eurlex_resources","dataset:joelito/legal-mc4","dataset:projecte-aina/CATalog","dataset:UFRGS/brwac","dataset:community-datasets/hrwac","dataset:danish-foundation-models/danish-gigaword","dataset:HiTZ/euscrawl","dataset:PleIAs/French-PD-Newspapers","dataset:PleIAs/French-PD-Books","dataset:AI-team-UoA/greek_legal_code","dataset:HiTZ/latxa-corpus-v1.1","dataset:allenai/peS2o","dataset:pile-of-law/pile-of-law","dataset:PORTULAN/parlamento-pt","dataset:hoskinson-center/proof-pile","dataset:togethercomputer/RedPajama-Data-1T","dataset:bigcode/starcoderdata","dataset:bjoernp/tagesschau-2018-2023","dataset:EleutherAI/the_pile_deduplicated","arxiv:2502.08489","arxiv:2403.14009","arxiv:2403.20266","arxiv:2101.00027","arxiv:2207.00220","arxiv:1810.06694","arxiv:1911.05507","arxiv:1906.03741","arxiv:2406.17557","arxiv:2402.06619","arxiv:1803.09010","base_model:BSC-LT/salamandra-7b","base_model:finetune:BSC-LT/salamandra-7b","license:apache-2.0","autotrain_compatible","text-generation-inference","endpoints_compatible","region:us"],"string":"[\n \"transformers\",\n \"safetensors\",\n \"llama\",\n \"text-generation\",\n \"conversational\",\n \"bg\",\n \"ca\",\n \"code\",\n \"cs\",\n \"cy\",\n \"da\",\n \"de\",\n \"el\",\n \"en\",\n \"es\",\n \"et\",\n \"eu\",\n \"fi\",\n \"fr\",\n \"ga\",\n \"gl\",\n \"hr\",\n \"hu\",\n \"it\",\n \"lt\",\n \"lv\",\n \"mt\",\n \"nl\",\n \"nn\",\n \"oc\",\n \"pl\",\n \"pt\",\n \"ro\",\n \"ru\",\n \"sh\",\n \"sk\",\n \"sl\",\n \"sr\",\n \"sv\",\n \"uk\",\n \"dataset:oscar-corpus/colossal-oscar-1.0\",\n \"dataset:HuggingFaceFW/fineweb-edu\",\n \"dataset:joelniklaus/eurlex_resources\",\n \"dataset:joelito/legal-mc4\",\n \"dataset:projecte-aina/CATalog\",\n \"dataset:UFRGS/brwac\",\n \"dataset:community-datasets/hrwac\",\n \"dataset:danish-foundation-models/danish-gigaword\",\n \"dataset:HiTZ/euscrawl\",\n \"dataset:PleIAs/French-PD-Newspapers\",\n \"dataset:PleIAs/French-PD-Books\",\n \"dataset:AI-team-UoA/greek_legal_code\",\n \"dataset:HiTZ/latxa-corpus-v1.1\",\n \"dataset:allenai/peS2o\",\n \"dataset:pile-of-law/pile-of-law\",\n \"dataset:PORTULAN/parlamento-pt\",\n \"dataset:hoskinson-center/proof-pile\",\n \"dataset:togethercomputer/RedPajama-Data-1T\",\n \"dataset:bigcode/starcoderdata\",\n \"dataset:bjoernp/tagesschau-2018-2023\",\n \"dataset:EleutherAI/the_pile_deduplicated\",\n \"arxiv:2502.08489\",\n \"arxiv:2403.14009\",\n \"arxiv:2403.20266\",\n \"arxiv:2101.00027\",\n \"arxiv:2207.00220\",\n \"arxiv:1810.06694\",\n \"arxiv:1911.05507\",\n \"arxiv:1906.03741\",\n \"arxiv:2406.17557\",\n \"arxiv:2402.06619\",\n \"arxiv:1803.09010\",\n \"base_model:BSC-LT/salamandra-7b\",\n \"base_model:finetune:BSC-LT/salamandra-7b\",\n \"license:apache-2.0\",\n \"autotrain_compatible\",\n \"text-generation-inference\",\n \"endpoints_compatible\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1727,"string":"1,727"},"last_modified":{"kind":"number","value":1740,"string":"1,740"},"downloads":{"kind":"number","value":12988,"string":"12,988"},"likes":{"kind":"number","value":54,"string":"54"},"README":{"kind":"string","value":"---\nbase_model:\n- BSC-LT/salamandra-7b\ndatasets:\n- oscar-corpus/colossal-oscar-1.0\n- HuggingFaceFW/fineweb-edu\n- joelniklaus/eurlex_resources\n- joelito/legal-mc4\n- projecte-aina/CATalog\n- UFRGS/brwac\n- community-datasets/hrwac\n- danish-foundation-models/danish-gigaword\n- HiTZ/euscrawl\n- PleIAs/French-PD-Newspapers\n- PleIAs/French-PD-Books\n- AI-team-UoA/greek_legal_code\n- HiTZ/latxa-corpus-v1.1\n- allenai/peS2o\n- pile-of-law/pile-of-law\n- PORTULAN/parlamento-pt\n- hoskinson-center/proof-pile\n- togethercomputer/RedPajama-Data-1T\n- bigcode/starcoderdata\n- bjoernp/tagesschau-2018-2023\n- EleutherAI/the_pile_deduplicated\nlanguage:\n- bg\n- ca\n- code\n- cs\n- cy\n- da\n- de\n- el\n- en\n- es\n- et\n- eu\n- fi\n- fr\n- ga\n- gl\n- hr\n- hu\n- it\n- lt\n- lv\n- mt\n- nl\n- nn\n- \\no\n- oc\n- pl\n- pt\n- ro\n- ru\n- sh\n- sk\n- sl\n- sr\n- sv\n- uk\nlibrary_name: transformers\nlicense: apache-2.0\npipeline_tag: text-generation\n---\n\n\n\n# Salamandra Model Card\n\nThis repository contains the model described in [Salamandra Technical Report](https://huggingface.co/papers/2502.08489).\n\nSalamandra is a highly multilingual model pre-trained from scratch that comes in three different \nsizes — 2B, 7B and 40B parameters — with their respective base and instruction-tuned variants. \nThis model card corresponds to the 7B instructed version.\n\nTo visit the model cards of other Salamandra versions, please refer to the [Model Index](#model-index).\n\nThe entire Salamandra family is released under a permissive [Apache 2.0 license](https://www.apache.org/licenses/LICENSE-2.0).\nAlong with the open weights, all training scripts and configuration files are made publicly available in [this GitHub repository](https://github.com/langtech-bsc/salamandra).\n\n> [!WARNING]\n> **DISCLAIMER:** This model is a first proof-of-concept designed to demonstrate the instruction-following capabilities of recently released base models.\n> It has been optimized to engage in conversation but has *NOT* been aligned through RLHF to filter or avoid sensitive topics.\n> As a result, it may generate harmful or inappropriate content.\n> The team is actively working to enhance its performance through further instruction and alignment with RL techniques.\n\n---\n\n## Model Details\n\n### Description\n\nTransformer-based decoder-only language model that has been pre-trained from scratch on 12.875 trillion tokens of highly curated data.\nThe pre-training corpus contains text in 35 European languages and code.\n\n### Hyperparameters\n\nThe full list of hyperparameters for each model can be found [here](https://github.com/langtech-bsc/salamandra/blob/main/configs/bsc_7b.yaml).\n\n### Architecture\n\n| | |\n|-------------------------|:--------------|\n| Total Parameters | 7,768,117,248 |\n| Embedding Parameters | 1,048,576,000 |\n| Layers | 32 |\n| Hidden size | 4,096 |\n| Attention heads | 32 |\n| Context length | 8,192 |\n| Vocabulary size | 256,000 |\n| Precision | bfloat16 |\n| Embedding type | RoPE |\n| Activation Function | SwiGLU |\n| Layer normalization | RMS Norm |\n| Flash attention | ✅ |\n| Grouped Query Attention | ✅ |\n| Num. query groups | 8 |\n\n---\n\n## Intended Use\n\n### Direct Use\n\nThe models are intended for both research and commercial use in any of the languages included in the training data. \nThe base models are intended either for language generation or to be further fine-tuned for specific use-cases. \nThe instruction-tuned variants can be used as general-purpose assistants, as long as the user is fully aware of the model’s limitations.\n\n### Out-of-scope Use\n\nThe model is not intended for malicious activities, such as harming others or violating human rights. \nAny downstream application must comply with current laws and regulations. \nIrresponsible usage in production environments without proper risk assessment and mitigation is also discouraged. \n\n---\n\n## Hardware and Software\n\n### Training Framework\n\nPre-training was conducted using NVIDIA’s [NeMo Framework](https://docs.nvidia.com/nemo-framework/index.html), \nwhich leverages PyTorch Lightning for efficient model training in highly distributed settings.\n\nThe instruction-tuned versions were produced with [FastChat](https://github.com/lm-sys/FastChat).\n\n### Compute Infrastructure\n\nAll models were trained on [MareNostrum 5](https://www.bsc.es/ca/marenostrum/marenostrum-5), a pre-exascale EuroHPC supercomputer hosted and\noperated by Barcelona Supercomputing Center.\n\nThe accelerated partition is composed of 1,120 nodes with the following specifications:\n- 4x Nvidia Hopper GPUs with 64GB HBM2 memory\n- 2x Intel Sapphire Rapids 8460Y+ at 2.3Ghz and 32c each (64 cores)\n- 4x NDR200 (BW per node 800Gb/s)\n- 512 GB of Main memory (DDR5)\n- 460GB on NVMe storage\n\n|Model|Nodes|GPUs|\n|:---:|:---:|:---:|\n|2B|64|256|\n|7B|128|512|\n|40B|256 / 512|1,024 / 2,048|\n\n---\n\n## How to use\n\nThe instruction-following models use the commonly adopted ChatML template:\n\n```jinja\n{%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content'] %}{%- set loop_messages = messages[1:] %}{%- else %}{%- set system_message = 'SYSTEM MESSAGE' %}{%- set loop_messages = messages %}{%- endif %}{%- if not date_string is defined %}{%- set date_string = '2024-09-30' %}{%- endif %}{{ '<|im_start|>system\\n' + system_message + '<|im_end|>\\n' }}{% for message in loop_messages %}{%- if (message['role'] != 'user') and (message['role'] != 'assistant')%}{{ raise_exception('Only user and assitant roles are suported after the initial optional system message.') }}{% endif %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{{'<|im_start|>' + message['role'] + '\\n' + message['content'] + '<|im_end|>' + '\\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\\n' }}{% endif %}\n```\nWhere `system_message` is used to guide the model during generation and `date_string` can be set to allow the model to respond with the current date.\n\nThe exact same chat template should be used for an enhanced conversational experience.\nThe easiest way to apply it is by using the tokenizer's built-in functions, as shown in the following snippet.\n\n```python\nfrom datetime import datetime\nfrom transformers import AutoTokenizer, AutoModelForCausalLM\nimport transformers\nimport torch\n\nmodel_id = \"BSC-LT/salamandra-7b-instruct\"\n\ntext = \"At what temperature does water boil?\"\n\ntokenizer = AutoTokenizer.from_pretrained(model_id)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_id,\n device_map=\"auto\",\n torch_dtype=torch.bfloat16\n )\n\nmessage = [ { \"role\": \"user\", \"content\": text } ]\ndate_string = datetime.today().strftime('%Y-%m-%d')\n\nprompt = tokenizer.apply_chat_template(\n message,\n tokenize=False,\n add_generation_prompt=True,\n date_string=date_string\n)\n\ninputs = tokenizer.encode(prompt, add_special_tokens=False, return_tensors=\"pt\")\noutputs = model.generate(input_ids=inputs.to(model.device), max_new_tokens=200)\n\nprint(tokenizer.decode(outputs[0], skip_special_tokens=True))\n```\nUsing this template, each turn is preceded by a `<|im_start|>` delimiter and the role of the entity \n(either `user`, for content supplied by the user, or `assistant` for LLM responses), and finished with the `<|im_end|>` token.\n\n---\n\n## Data\n\n### Pretraining Data\n\nThe pre-training corpus comprises data from 35 European languages and 92 programming languages, with detailed data sources provided below. \nThe initial three training epochs used 2.4 trillion tokens, obtained by manually adjusting data proportion to balance the representation \nand give more importance to Spain’s co-official (Spanish, Catalan, Galician, and Basque). This way, we downsampled code and English data to half, \nSpanish co-official languages were oversampled by 2x, and the remaining languages were kept in their original proportions.\nDuring the following epochs, the Colossal OSCAR dataset was replaced with the FineWeb-Edu dataset. \nThis adjustment resulted in a total of 2.68 trillion tokens, distributed as outlined below:\n\n\n\nThe pretraining corpus is predominantly composed of data from Colossal OSCAR, which contributes a significant 53.05% of the total tokens. \nFollowing this, Starcoder provides 13.67%, and FineWeb-Edu (350BT subset) adds 10.24%. The next largest sources are HPLT at 4.21% and French-PD at 3.59%. \nOther notable contributions include MaCoCu, Legal-ES, and EurLex, each contributing around 1.72% to 1.41%. \nThese major sources collectively form the bulk of the corpus, ensuring a rich and diverse dataset for training the language model. \nThe remaining 10% comes from smaller sources in various languages.\n\nFeel free to click the expand button below to see the full list of sources.\n\n<details>\n<summary>Data Sources</summary>\n \n\n| Dataset | Language | Source |\n|---|---|---|\n| Colossal OSCAR 1.0 | bg, ca, cs, cy, da, de, el, en, es, et, eu, fi, fr, ga, gl, hr, hu, it, lt, lv, mt, nl, nn, no, oc, pl, pt, ro, ru, sh, sk, sl, sr, sv, uk | Brack et al., 2024 |\n| Aya Dataset (w/o Evaluation Suite) | eu, hr, nl, fi, ka, hu, lt, nn, ro, sk, lv, cy, bg, cs, en, fr, de, ga, mt, pl, ru, sl, sv, ca, da, et, gl, el, it, no, pt, sr, es, uk | Singh et al., 2024 |\n| Wikimedia dumps | bg, ca, cs, da, de, el, en, es, et, eu, fi, fr, ga, gl, hr, hu, it, lt, lv, mt, nl, nn, no, pl, pt, ro, sh, sk, sl, sr, uk | [Link](https://dumps.wikimedia.org/) |\n| OpenSubtitles v2016 | bg, ca, cs, da, de, el, en, es, et, eu, fi, fr, gl, hr, it, lt, lv, nl, no, pl, pt, ro, sk, sl, sr, sv, uk | Lison & Tiedemann, 2016 |\n| EurLEX-Resources | bg, cs, da, de, el, en, es, et, fi, fr, ga, hr, hu, it, lt, lv, mt, nl, pl, pt, ro, sk, sl, sv | [Link](https://huggingface.co/datasets/joelniklaus/eurlex_resources) |\n| MC4-Legal | bg, cs, da, de, el, en, es, et, fi, fr, ga, hu, it, lt, lv, mt, nl, pl, pt, ro, sk, sl, sv | [Link](https://huggingface.co/datasets/joelito/legal-mc4) |\n| Parlamint | at, bg, cz, dk, ee, es, es-ga, fi, fr, gb, gr, hr, hu, it, lv, nl, no, pl, pt, rs, se, si | Erjavec et al., 2021 |\n| MaCoCu | bg, ca, el, hr, mt, sl, sr, uk | Bañón et al., 2022 |\n| CURLICAT | bg, hr, hu, pl, ro, sk, sl | Váradi et al., 2022 |\n| Norwegian Colossal Corpus (NCC) | nn, no | Kummervold et al., 2021 |\n| Academic Slovene KAS 2.0 | sl | Žagar et al., 2022 |\n| BIGPATENT | en | Sharma et al., 2019 |\n| Biomedical-ES | es | Internally generated biomedical dataset: Wikipedia LS, Pubmed, MeSpEn, patents, clinical cases, medical crawler |\n| Brazilian Portuguese Web as Corpus (BrWaC) | pt | Wagner Filho et al., 2018 |\n| Bulgarian National Corpus (BulNC) | bg | [Link](http://old.dcl.bas.bg/dataset/BulNC.7z) |\n| CaBeRnet | fr | Popa-Fabre et al., 2020 |\n| CATalog 1.0 | ca | Palomar-Giner et al., 2024 |\n| CorpusNÓS | gl | de-Dios-Flores et al., 2024 |\n| Croatian Web as Corpus 2.1 (hrWaC) | hr | Ljubešić & Klubička, 2014 |\n| DaNewsroom | da | Varab & Schluter, 2020 |\n| Danish GigaWord | da | Strømberg-Derczynski et al., 2021 |\n| DK-CLARIN Reference Corpus of General Danish | da | [Link](https://korpus.dsl.dk/clarin/) |\n| Estonian National Corpus 2021 (ENC) | et | Koppel & Kallas, 2022 |\n| Estonian Reference Corpus (ERC) | et | [Link](https://www.cl.ut.ee/korpused/segakorpus/) |\n| EusCrawl (w/o Wikipedia or NC-licenses) | eu | Artetxe et al., 2022 |\n| FineWeb-Edu (350BT subset) | en | Penedo et al., 2024 |\n| French Public Domain Books (French-PD) | fr | [Link](https://huggingface.co/datasets/PleIAs/French-PD-Books) |\n| French Public Domain Newspapers (French-PD) | fr | [Link](https://huggingface.co/datasets/PleIAs/French-PD-Newspapers) |\n| German Web as Corpus (DeWaC) | de | [Link](https://docs.sslmit.unibo.it/doku.php?id=corpora:dewac) |\n| Greek Legal Code (GLC) | el | Papaloukas et al., 2021 |\n| Greek Web Corpus (GWC) | el | Outsios et al., 2018 |\n| HPLT v1 - Spanish | es | de Gibert et al., 2024 |\n| HPLT v1.1 - Spanish | es | de Gibert et al., 2024 |\n| Irish Universal Dependencies (Ga-UD) | ga | [Link](https://universaldependencies.org/ga/index.html) |\n| Italian Web as Corpus (ItWaC) | it | [Link](https://docs.sslmit.unibo.it/doku.php?id=corpora:itwac) |\n| Korpus Malti | mt | Micallef et al., 2022 |\n| Korpus slovenských právnych predpisov v1.9 (SK-Laws) | sk | [Link](https://www.juls.savba.sk/data/marcell/legal-sk-20220322-1.9.ver.xz) |\n| Latxa Corpus v1.1 (GAITU) | eu | Etxaniz et al., 2024 [Link](https://huggingface.co/datasets/HiTZ/latxa-corpus-v1.1) |\n| Laws and legal acts of Ukraine (UK-Laws) | uk | [Link](https://lang.org.ua/en/corpora/#anchor7) |\n| Legal-ES | es | Internally generated legal dataset: BOE, BORME, Senado, Congreso, Spanish court orders, DOGC |\n| MARCELL Romanian legislative subcorpus v2 | ro | [Link](https://elrc-share.eu/reposMARCELL%20Romanian%20legislative%20subcorpus%20v2itory/browse/marcell-romanian-legislative-subcorpus-v2/2da548428b9d11eb9c1a00155d026706ce94a6b59ffc4b0e9fb5cd9cebe6889e/) |\n| Math AMPS | en | Hendrycks et al., 2021 |\n| NKPJ National Corpus of Polish v1.2 (NKPJ) | pl | Lewandowska-Tomaszczyk et al., 2013 |\n| Occitan Corpus (IEA-AALO) | oc | Provided by [IEA](https://www.institutestudisaranesi.cat/) |\n| Open Legal Data - German court decisions and laws | de | Ostendorff et al., 2020 |\n| ParlamentoPT | pt | Rodrigues et al., 2023 |\n| peS2o | en | Soldaini & Lo, 2023 |\n| PG-19 | en | Rae et al., 2019 |\n| Pile of Law (selected subsets) | en | Henderson* et al., 2022 |\n| Polish Parliamentary Corpus (PPC) | pl | Ogrodniczuk, 2018 |\n| Proof Pile | en | [Link](https://huggingface.co/datasets/hoskinson-center/proof-pile) |\n| RedPajama-Data T1 (StackExchange subset) | en | Computer, 2023 |\n| Scientific-ES | es | Internally generated scientific dataset: Dialnet, Scielo, CSIC, TDX, BSC, UCM |\n| SK Court Decisions v2.0 (OD-Justice) | sk | [Link](https://www.juls.savba.sk/data/od-justice/od-justice-2.0.ver.xz) |\n| Slovene Web as Corpus (slWaC) | sl | Erjavec et al., 2015 |\n| SoNaR Corpus NC 1.2 | nl | [Link](https://taalmaterialen.ivdnt.org/download/tstc-sonar-corpus/) |\n| Spanish Legal Domain Corpora (Spanish-Legal) | es | Gutiérrez-Fandiño et al., 2021 |\n| SrpKorSubset: news, legal, academic, conversation, lit- erary (SrpKor) | sr | [Link](http://www.korpus.matf.bg.ac.rs/) |\n| Starcoder | code | Li et al., 2023 |\n| State-related content from the Latvian Web (State-Latvian-Web) | lv | [Link](https://catalog.elra.info/en-us/repository/browse/ELRA-W0169/) |\n| SYN v9: large corpus of written Czech | cs | Křen et al., 2021 |\n| Tagesschau Archive Article | de | [Link](https://huggingface.co/datasets/bjoernp/tagesschau-2018-2023) |\n| The Danish Parliament Corpus 2009 - 2017, v1 | da | Hansen, 2018 |\n| The Gaois bilingual corpus of English-Irish legislation (Ga-Legislation) | ga | [Link](https://portulanclarin.net/repository/browse/the-gaois-bilingual-corpus-of-english-irish-legislation-processed/daeac17c9e3511ea9b7f02420a000407b83de243dc0b469aab41084386c5b80f/) |\n| The Pile (PhilPapers) | en | Gao et al., 2021 |\n| The Swedish Culturomics Gigaword Corpus (Swedish- Gigaword) | sv | Rødven-Eide, 2016 |\n| Welsh-GOV | cy | Crawling from [Link](https://www.llyw.cymru) |\n| Yle Finnish News Archive (Yle-News) | fi | [Link](http://urn.fi/urn:nbn:fi:lb-2021050401) |\n\nTo consult the data summary document with the respective licences, please send an e-mail to ipr@bsc.es. \n\n<details>\n<summary>References</summary>\n\n- Abadji, J., Suárez, P. J. O., Romary, L., & Sagot, B. (2021). Ungoliant: An optimized pipeline for the generation of a very large-scale multilingual web corpus (H. Lüngen, M. Kupietz, P. Bański, A. Barbaresi, S. Clematide, & I. Pisetta, Eds.; pp. 1–9). Leibniz-Institut für Deutsche Sprache. [Link](https://doi.org/10.14618/ids-pub-10468)\n- Artetxe, M., Aldabe, I., Agerri, R., Perez-de-Viñaspre, O., & Soroa, A. (2022). Does Corpus Quality Really Matter for Low-Resource Languages?\n- Bañón, M., Esplà-Gomis, M., Forcada, M. L., García-Romero, C., Kuzman, T., Ljubešić, N., van Noord, R., Sempere, L. P., Ramírez-Sánchez, G., Rupnik, P., Suchomel, V., Toral, A., van der Werff, T., & Zaragoza, J. (2022). MaCoCu: Massive collection and curation of monolingual and bilingual data: Focus on under-resourced languages. Proceedings of the 23rd Annual Conference of the European Association for Machine Translation, 303–304. [Link](https://aclanthology.org/2022.eamt-1.41)\n- Brack, M., Ostendorff, M., Suarez, P. O., Saiz, J. J., Castilla, I. L., Palomar-Giner, J., Shvets, A., Schramowski, P., Rehm, G., Villegas, M., & Kersting, K. (2024). Community OSCAR: A Community Effort for Multilingual Web Data. [Link](https://occiglot.eu/papers/Community_Oscar.pdf)\n- Computer, T. (2023). RedPajama: An Open Source Recipe to Reproduce LLaMA training dataset [Computer software]. [Link](https://github.com/togethercomputer/RedPajama-Data)\n- de Gibert, O., Nail, G., Arefyev, N., Bañón, M., van der Linde, J., Ji, S., Zaragoza-Bernabeu, J., Aulamo, M., Ramírez-Sánchez, G., Kutuzov, A., Pyysalo, S., Oepen, S., & Tiedemann, J. (2024). A New Massive Multilingual Dataset for High-Performance Language Technologies (arXiv:2403.14009). arXiv. [Link](http://arxiv.org/abs/2403.14009)\n- Dodge, J., Sap, M., Marasović, A., Agnew, W., Ilharco, G., Groeneveld, D., Mitchell, M., & Gardner, M. (2021). Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus. In M.-F. Moens, X. Huang, L. Specia, & S. W. Yih (Eds.), Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (pp. 1286–1305). Association for Computational Linguistics. [Link](https://doi.org/10.18653/v1/2021.emnlp-main.98)\n- Erjavec, T., Ljubešić, N., & Logar, N. (2015). The slWaC corpus of the Slovene web. Informatica (Slovenia), 39, 35–42.\n- Erjavec, T., Ogrodniczuk, M., Osenova, P., Ljubešić, N., Simov, K., Grigorova, V., Rudolf, M., Pančur, A., Kopp, M., Barkarson, S., Steingrímsson, S. hór, van der Pol, H., Depoorter, G., de Does, J., Jongejan, B., Haltrup Hansen, D., Navarretta, C., Calzada Pérez, M., de Macedo, L. D., … Rayson, P. (2021). Linguistically annotated multilingual comparable corpora of parliamentary debates ParlaMint.ana 2.1. [Link](http://hdl.handle.net/11356/1431)\n- Etxaniz, J., Sainz, O., Perez, N., Aldabe, I., Rigau, G., Agirre, E., Ormazabal, A., Artetxe, M., & Soroa, A. (2024). Latxa: An Open Language Model and Evaluation Suite for Basque. [Link] (https://arxiv.org/abs/2403.20266)\n- Gao, L., Biderman, S., Black, S., Golding, L., Hoppe, T., Foster, C., Phang, J., He, H., Thite, A., Nabeshima, N., Presser, S., & Leahy, C. (2021). The Pile: An 800GB Dataset of Diverse Text for Language Modeling. CoRR, abs/2101.00027. [Link](https://arxiv.org/abs/2101.00027)\n- Gutiérrez-Fandiño, A., Armengol-Estapé, J., Gonzalez-Agirre, A., & Villegas, M. (2021). Spanish Legalese Language Model and Corpora.\n- Hansen, D. H. (2018). The Danish Parliament Corpus 2009—2017, v1. [Link](http://hdl.handle.net/20.500.12115/8)\n- Henderson*, P., Krass*, M. S., Zheng, L., Guha, N., Manning, C. D., Jurafsky, D., & Ho, D. E. (2022). Pile of Law: Learning Responsible Data Filtering from the Law and a 256GB Open-Source Legal Dataset. arXiv. [Link](https://arxiv.org/abs/2207.00220)\n- Hendrycks, D., Burns, C., Kadavath, S., Arora, A., Basart, S., Tang, E., Song, D., & Steinhardt, J. (2021). Measuring Mathematical Problem Solving With the MATH Dataset. NeurIPS.\n- Jansen, T., Tong, Y., Zevallos, V., & Suarez, P. O. (2022). Perplexed by Quality: A Perplexity-based Method for Adult and Harmful Content Detection in Multilingual Heterogeneous Web Data.\n- Koppel, K., & Kallas, J. (2022). Eesti keele ühendkorpuste sari 2013–2021: Mahukaim eestikeelsete digitekstide kogu. Eesti Rakenduslingvistika Ühingu Aastaraamat Estonian Papers in Applied Linguistics, 18, 207–228. [Link](https://doi.org/10.5128/erya18.12)\n- Křen, M., Cvrček, V., Henyš, J., Hnátková, M., Jelínek, T., Kocek, J., Kováříková, D., Křivan, J., Milička, J., Petkevič, V., Procházka, P., Skoumalová, H., Šindlerová, J., & Škrabal, M. (2021). SYN v9: Large corpus of written Czech. [Link](http://hdl.handle.net/11234/1-4635)\n- Kreutzer, J., Caswell, I., Wang, L., Wahab, A., van Esch, D., Ulzii-Orshikh, N., Tapo, A., Subramani, N., Sokolov, A., Sikasote, C., Setyawan, M., Sarin, S., Samb, S., Sagot, B., Rivera, C., Rios, A., Papadimitriou, I., Osei, S., Suarez, P. O., … Adeyemi, M. (2022). Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets. Transactions of the Association for Computational Linguistics, 10, 50–72. [Link](https://doi.org/10.1162/tacl_a_00447)\n- Kummervold, P. E., De la Rosa, J., Wetjen, F., & Brygfjeld, S. A. (2021). Operationalizing a National Digital Library: The Case for a Norwegian Transformer Model. In S. Dobnik & L. Øvrelid (Eds.), Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa) (pp. 20–29). Linköping University Electronic Press, Sweden. [Link](https://aclanthology.org/2021.nodalida-main.3)\n- Lewandowska-Tomaszczyk, B., Górski, R., Łaziński, M., & Przepiórkowski, A. (2013). The National Corpus of Polish (NKJP). Language use and data analysis. 309–319.\n- Li, R., Allal, L. B., Zi, Y., Muennighoff, N., Kocetkov, D., Mou, C., Marone, M., Akiki, C., Li, J., Chim, J., Liu, Q., Zheltonozhskii, E., Zhuo, T. Y., Wang, T., Dehaene, O., Davaadorj, M., Lamy-Poirier, J., Monteiro, J., Shliazhko, O., … Vries, H. de. (2023). StarCoder: May the source be with you!\n- Lison, P., & Tiedemann, J. (2016). OpenSubtitles2016: Extracting Large Parallel Corpora from Movie and TV Subtitles. In N. Calzolari, K. Choukri, T. Declerck, S. Goggi, M. Grobelnik, B. Maegaard, J. Mariani, H. Mazo, A. Moreno, J. Odijk, & S. Piperidis (Eds.), Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16) (pp. 923–929). European Language Resources Association (ELRA). [Link](https://aclanthology.org/L16-1147)\n- Ljubešić, N., & Klubička, F. (2014). Bs,hr,srWaC - Web Corpora of Bosnian, Croatian and Serbian. In F. Bildhauer & R. Schäfer (Eds.), Proceedings of the 9th Web as Corpus Workshop (WaC-9) (pp. 29–35). Association for Computational Linguistics. [Link](https://doi.org/10.3115/v1/W14-0405)\n- Micallef, K., Gatt, A., Tanti, M., van der Plas, L., & Borg, C. (2022). Pre-training Data Quality and Quantity for a Low-Resource Language: New Corpus and BERT Models for Maltese. Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing, 90–101. [Link](https://doi.org/10.18653/v1/2022.deeplo-1.10)\n- Ogrodniczuk, M. (2018). Polish Parliamentary Corpus. [Link](https://api.semanticscholar.org/CorpusID:235134113)\n- Ostendorff, M., Blume, T., & Ostendorff, S. (2020). Towards an Open Platform for Legal Information. Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020, 385–388. [Link](https://doi.org/10.1145/3383583.3398616)\n- Ostendorff, M., Suarez, P. O., Lage, L. F., & Rehm, G. (2024). LLM-Datasets: An Open Framework for Pretraining Datasets of Large Language Models. First Conference on Language Modeling. [Link](https://openreview.net/forum?id=5RdIMlGLXL)\n- Outsios, S., Skianis, K., Meladianos, P., Xypolopoulos, C., & Vazirgiannis, M. (2018). Word Embeddings from Large-Scale Greek Web content. arXiv Preprint arXiv:1810.06694.\n- Palomar-Giner, J., Saiz, J. J., Espuña, F., Mina, M., Da Dalt, S., Llop, J., Ostendorff, M., Ortiz Suarez, P., Rehm, G., Gonzalez-Agirre, A., & Villegas, M. (2024). A CURATEd CATalog: Rethinking the Extraction of Pretraining Corpora for Mid-Resourced Languages. In N. Calzolari, M.-Y. Kan, V. Hoste, A. Lenci, S. Sakti, & N. Xue (Eds.), Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024) (pp. 335–349). ELRA and ICCL. [Link](https://aclanthology.org/2024.lrec-main.31)\n- Papaloukas, C., Chalkidis, I., Athinaios, K., Pantazi, D.-A., & Koubarakis, M. (2021). Multi-granular Legal Topic Classification on Greek Legislation. Proceedings of the Natural Legal Language Processing Workshop 2021, 63–75. [Link](https://doi.org/10.48550/arXiv.2109.15298)\n- Popa-Fabre, M., Ortiz Suárez, P. J., Sagot, B., & de la Clergerie, É. (2020). French Contextualized Word-Embeddings with a sip of CaBeRnet: A New French Balanced Reference Corpus. Proceedings of the 8th Workshop on Challenges in the Management of Large Corpora, 15–23. [Link](https://aclanthology.org/2020.cmlc-1.3)\n- Rae, J. W., Potapenko, A., Jayakumar, S. M., Hillier, C., & Lillicrap, T. P. (2019). Compressive Transformers for Long-Range Sequence Modelling. arXiv Preprint. [Link](https://arxiv.org/abs/1911.05507)\n- Rodrigues, J., Gomes, L., Silva, J., Branco, A., Santos, R., Cardoso, H. L., & Osório, T. (2023). Advancing Neural Encoding of Portuguese with Transformer Albertina PT-\\*.\n- Rødven-Eide, S. (2016). The Swedish Culturomics Gigaword CorpusThe Swedish Culturomics Gigaword Corpus [Dataset]. Språkbanken Text. [Link](https://doi.org/10.23695/3WMV-1Z09)\n- Sharma, E., Li, C., & Wang, L. (2019). BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization. CoRR, abs/1906.03741. [Link](http://arxiv.org/abs/1906.03741)\n- Soldaini, L., & Lo, K. (2023). peS2o (Pretraining Efficiently on S2ORC) Dataset. Allen Institute for AI.\n- Strømberg-Derczynski, L., Ciosici, M., Baglini, R., Christiansen, M. H., Dalsgaard, J. A., Fusaroli, R., Henrichsen, P. J., Hvingelby, R., Kirkedal, A., Kjeldsen, A. S., Ladefoged, C., Nielsen, F. Å., Madsen, J., Petersen, M. L., Rystrøm, J. H., & Varab, D. (2021). The Danish Gigaword Corpus. Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa), 413–421. [Link](https://aclanthology.org/2021.nodalida-main.46)\n- Subramani, N., Luccioni, S., Dodge, J., & Mitchell, M. (2023). Detecting Personal Information in Training Corpora: An Analysis. 208–220. [Link](https://doi.org/10.18653/v1/2023.trustnlp-1.18)\n- Varab, D., & Schluter, N. (2020). DaNewsroom: A Large-scale Danish Summarisation Dataset. Proceedings of The 12th Language Resources and Evaluation Conference, 6731–6739. [Link](https://www.aclweb.org/anthology/2020.lrec-1.831)\n- Váradi, T., Nyéki, B., Koeva, S., Tadić, M., Štefanec, V., Ogrodniczuk, M., Nitoń, B., Pezik, P., Barbu Mititelu, V., Irimia, E., Mitrofan, M., Tufi\\textcommabelows, D., Garabík, R., Krek, S., & Repar, A. (2022). Introducing the CURLICAT Corpora: Seven-language Domain Specific Annotated Corpora from Curated Sources. In N. Calzolari, F. Béchet, P. Blache, K. Choukri, C. Cieri, T. Declerck, S. Goggi, H. Isahara, B. Maegaard, J. Mariani, H. Mazo, J. Odijk, & S. Piperidis (Eds.), Proceedings of the Thirteenth Language Resources and Evaluation Conference (pp. 100–108). European Language Resources Association. [Link](https://aclanthology.org/2022.lrec-1.11)\n- Wagner Filho, J. A., Wilkens, R., Idiart, M., & Villavicencio, A. (2018). The brwac corpus: A new open resource for brazilian portuguese. Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018).\n- Žagar, A., Kavaš, M., Robnik-Šikonja, M., Erjavec, T., Fišer, D., Ljubešić, N., Ferme, M., Borovič, M., Boškovič, B., Ojsteršek, M., & Hrovat, G. (2022). Corpus of academic Slovene KAS 2.0. [Link](http://hdl.handle.net/11356/1448)\n- Alicia Parrish, Angelica Chen, Nikita Nangia, Vishakh Padmakumar, Jason Phang, Jana Thompson, Phu Mon Htut, and Samuel Bowman. 2022. BBQ: A hand-built bias benchmark for question answering. In Findings of the Association for Computational Linguistics: ACL 2022, pages 2086–2105, Dublin, Ireland. Association for Computational Linguistics.\n- Emily Sheng, Kai-Wei Chang, Premkumar Natarajan, and Nanyun Peng. 2019. The Woman Worked as a Babysitter: On Biases in Language Generation. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3407–3412, Hong Kong, China. Association for Computational Linguistics.\n- Clark, P., Cowhey, I., Etzioni, O., Khot, T., Sabharwal, A., Schoenick, C., & Tafjord, O. (2018). Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge. arXiv:1803. 05457v1.\n- Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D. Manning, Andrew Ng, and Christopher Potts. 2013. Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1631–1642, Seattle, Washington, USA. Association for Computational Linguistics.\n- Penedo, G., Kydlíček, H., allal, L. B., Lozhkov, A., Mitchell, M., Raffel, C., Von Werra, L., & Wolf, T. (2024). The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale (arXiv:2406.17557). arXiv. http://arxiv.org/abs/2406.17557\n- Singh, S., Vargus, F., Dsouza, D., Karlsson, B. F., Mahendiran, A., Ko, W.-Y., Shandilya, H., Patel, J., Mataciunas, D., OMahony, L., Zhang, M., Hettiarachchi, R., Wilson, J., Machado, M., Moura, L. S., Krzemiński, D., Fadaei, H., Ergün, I., Okoh, I., … Hooker, S. (2024). Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning (arXiv:2402.06619). arXiv. http://arxiv.org/abs/2402.06619\n\n</details>\n\n</details>\n\nThe model was trained on 3 pre-training epochs with 2.4T tokens per epoch, 2 additional pre-training epochs in which the English part \nof the Colossal OSCAR dataset was replaced with FineWeb-Edu (350BT subset), resulting in 2.68T tokens per epoch; \nand 1 final epoch of 0.315T higher quality tokens, meaning that the total number of tokens seen during pre-training is approximately 12.875 trillion tokens.\n\nWe provide an extense Datasheet section following the best practices defined by [(Gebru et al., 2021)](https://arxiv.org/pdf/1803.09010).\n\n<details>\n<summary>Datasheet</summary>\n\n#### Motivation\n\n**For what purpose was the dataset created? Was there a specific task in mind? Was there a specific gap that needed to be filled? Please provide a description.**\n\nThe purpose of creating this dataset is to pre-train the Salamandra family of multilingual models with high performance in a large number of European languages (35) \nand programming languages (92). We also want to represent the co-official languages of Spain: Spanish, Catalan, Galician and Basque. For this reason, we oversample \nthese languages by a factor of 2.\n\nThere is a great lack of massive multilingual data, especially in minority languages (Ostendorff & Rehm, 2023), so part of our efforts in the creation of \nthis pre-training dataset have resulted in the contribution to large projects such as the Community OSCAR (Brack et al., 2024), which includes 151 languages \nand 40T words, or CATalog (Palomar-Giner et al., 2024), the largest open dataset in Catalan in the world.\n\n**Who created the dataset (e.g., which team, research group) and on behalf of which entity (e.g., company, institution, organization)?**\n\nThe dataset has been created by the Language Technologies unit (LangTech) of the Barcelona Supercomputing Center - Centro Nacional de Supercomputación (BSC-CNS), \nwhich aims to advance the field of natural language processing through cutting-edge research and development and the use of HPC. In particular, it was created by \nthe unit's data team, the main contributors being José Javier Saiz, Ferran Espuña and Jorge Palomar.\n\nHowever, the creation of the dataset would not have been possible without the collaboration of a large number of collaborators, partners and public institutions, \nwhich can be found in detail in the acknowledgements.\n\n**Who funded the creation of the dataset? If there is an associated grant, please provide the name of the grantor and the grant name and number.**\n\nThis work has been promoted and financed by the Government of Catalonia through the [Aina project](https://projecteaina.cat/).\n\nThis work is funded by the _Ministerio para la Transformación Digital y de la Función Pública_ - Funded by EU – NextGenerationEU \nwithin the framework of [ILENIA Project](https://proyectoilenia.es/) with reference 2022/TL22/00215337.\n\n#### Composition\n\n**What do the instances that comprise the dataset represent (e.g., documents, photos, people, countries)? Are there multiple types of instances (e.g., movies, users, and ratings; people and interactions between them; nodes and edges)? Please provide a description.**\n\nThe dataset consists entirely of text documents in various languages. Specifically, data was mainly sourced from the following databases and \nrepositories:\n- **Common Crawl:** Repository that holds website data and is run by the Common Crawl non-profit organization. It is updated monthly and is\n distributed under the CC0 1.0 public domain license.\n- **GitHub:** Community platform that allows developers to create, store, manage, and share their code. Repositories are crawled and then\n distributed with their original licenses, which may vary from permissive to non-commercial licenses.\n- **Wikimedia:** Database that holds the collection databases managed by the Wikimedia Foundation, including Wikipedia, Wikibooks, Wikinews,\n Wikiquote, Wikisource, and Wikivoyage. It is updated monthly and is distributed under Creative Commons Attribution-ShareAlike License 4.0.\n- **EurLex:** Repository that holds the collection of legal documents from the European Union, available in all of the EU’s 24 official\n languages and run by the Publications Office of the European Union. It is updated daily and is distributed under the Creative Commons\n Attribution 4.0 International license.\n- **Other repositories:** Specific repositories were crawled under permission for domain-specific corpora, which include academic, legal,\n and newspaper repositories.\n\nWe provide a complete list of dataset sources at the end of this section.\n\n**How many instances are there in total (of each type, if appropriate)?**\n\nThe dataset contains a diverse range of instances across multiple languages, with notable adjustments for certain languages. English \nrepresents the largest portion, accounting for 39.31% of the total data. Spanish was upsampled by a factor of 2, bringing its share to 16.12%,\nwhile Catalan (1.97%), Basque (0.24%), and Galician (0.31%) were also upsampled by 2. On the other hand, code-related data was downsampled \nby half, making up 5.78% of the total. Other prominent languages include French (6.6%), Russian (5.56%), German (4.79%), and Hungarian \n(4.59%), with several additional languages contributing between 1% and 2%, and smaller portions represented by a variety of others.\n\n**Does the dataset contain all possible instances or is it a sample (not necessarily random) of instances from a larger set? If the dataset is a sample, then what is the larger set? Is the sample representative of the larger set (e.g., geographic coverage)? If so, please describe how this representativeness was validated/verified. If it is not representative of the larger set, please describe why not (e.g., to cover a more diverse range of instances, because instances were withheld or unavailable).**\n\nThe dataset is a sample from multiple sources, with different weights based on the primary language of the content: Spanish, Catalan, \nBasque, and Galician content was upsampled by a factor of two, while programming languages were downsampled by a factor of half. Other \nsources were sampled in proportion to their occurrence.\n\n**What data does each instance consist of? “Raw” data (e.g., unprocessed text or images) or features? In either case, please provide a description.**\n\nEach instance consists of a text document processed for deduplication, language identification, and source-specific filtering. Some documents required \noptical character recognition (OCR) to extract text from non-text formats such as PDFs.\n\n**Is there a label or target associated with each instance? If so, please provide a description.**\n\nEach instance is labelled with a unique identifier, the primary language of the content, and the URL for web-sourced instances. Additional labels were \nautomatically assigned to detect specific types of content -harmful or toxic content- and to assign preliminary indicators of undesired qualities -very \nshort documents, high density of symbols, etc.- which were used for filtering instances.\n\n**Is any information missing from individual instances? If so, please provide a description, explaining why this information is missing (e.g., because it was unavailable). This does not include intentionally removed information, but might include, e.g., redacted text.**\n\nNo significant information is missing from the instances.\n\n**Are relationships between individual instances made explicit (e.g., users’ movie ratings, social network links)? If so, please describe how these relationships are made explicit.**\n\nInstances are related through shared metadata, such as source and language identifiers.\n\n**Are there recommended data splits (e.g., training, development/validation, testing)? If so, please provide a description of these splits, explaining the rationale behind them.**\n\nThe dataset is randomly divided into training, validation and test sets, where the validation and test sets are each 1% of the total corpus.\n\n**Are there any errors, sources of noise, or redundancies in the dataset? If so, please provide a description.**\n\nDespite removing duplicated instances within each source, redundancy remains at the paragraph and sentence levels, particularly in web-sourced \ninstances where search engine optimization techniques and templates contribute to repeated textual patterns. Some instances may be also duplicated \nacross sources due to format variations.\n\n**Is the dataset self-contained, or does it link to or otherwise rely on external resources (e.g., websites, tweets, other datasets)? If it links to or relies on external resources, a) are there guarantees that they will exist, and remain constant, over time; b) are there official archival versions of the complete dataset (i.e., including the external resources as they existed at the time the dataset was created); c) are there any restrictions (e.g., licenses, fees) associated with any of the external resources that might apply to a dataset consumer? Please provide descriptions of all external resources and any restrictions associated with them, as well as links or other access points, as appropriate.**\n\nThe dataset is self-contained and does not rely on external resources.\n\n**Does the dataset contain data that might be considered confidential (e.g., data that is protected by legal privilege or by doctor–patient confidentiality, data that includes the content of individuals’ non-public communications)? If so, please provide a description.**\n\nThe dataset does not contain confidential data.\n\n**Does the dataset contain data that, if viewed directly, might be offensive, insulting, threatening, or might otherwise cause anxiety? If so, please describe why. If the dataset does not relate to people, you may skip the remaining questions in this section.**\n\nThe dataset includes web-crawled content, which may overrepresent pornographic material across languages (Kreutzer et al., 2022). Although \npre-processing techniques were applied to mitigate offensive content, the heterogeneity and scale of web-sourced data make exhaustive \nfiltering challenging, which makes it next to impossible to identify all adult content without falling into excessive filtering, which may \nnegatively influence certain demographic groups (Dodge et al., 2021).\n\n**Does the dataset identify any subpopulations (e.g., by age, gender)? If so, please describe how these subpopulations are identified and provide a description of their respective distributions within the dataset.**\n\nThe dataset does not explicitly identify any subpopulations.\n\n**Is it possible to identify individuals (i.e., one or more natural persons), either directly or indirectly (i.e., in combination with other data) from the dataset? If so, please describe how.**\n\nWeb-sourced instances in the dataset may contain personally identifiable information (PII) that is publicly available on the Web, such as names, \nIP addresses, email addresses, and phone numbers. While it would be possible to indirectly identify individuals through the combination of multiple \ndata points, the nature and scale of web data makes it difficult to parse such information. In any case, efforts are made to filter or anonymize \nsensitive data (Mina et al., 2024), but some identifiable information may remain in the dataset.\n\n**Does the dataset contain data that might be considered sensitive in any way? If so, please provide a description.**\n\nGiven that the dataset includes web-sourced content and other publicly available documents, instances may inadvertently reveal financial \ninformation, health-related details, or forms of government identification, such as social security numbers (Subramani et al., 2023), \nespecially if the content originates from less-regulated sources or user-generated platforms. \n\n#### Collection Process\n\n**How was the data collected?**\n\nThis dataset is constituted by combining several sources, whose acquisition methods can be classified into three groups:\n- Web-sourced datasets with some preprocessing available under permissive license.\n- Domain-specific or language-specific raw crawls.\n- Manually curated data obtained through collaborators, data providers (by means of legal assignment agreements) or open source projects (e.g. CATalog).\n\n**What mechanisms or procedures were used to collect the data? How were these mechanisms or procedures validated?**\n\nThe data collection process was carried out using three different mechanisms, each corresponding to one of the groups defined in the previous answer. The specific methods used and their respective validation procedures are outlined below:\n- Open Direct Download: Data were obtained directly from publicly accessible sources, such as websites or repositories that provide open data downloads. We validate the data with a data integrity check, which ensures that the downloaded files are complete, uncorrupted and in the expected format and structure.\n- Ad hoc scrapers or crawlers: Custom web scraping scripts or crawlers were used to extract data from various online sources where direct downloads were not available. These scripts navigate web pages, extract relevant data and store it in a structured format. We validate this method with software unit tests to evaluate the functionality of individual components of the scraping programs, checking for errors or unexpected behaviour. In addition, data integrity tests were performed to verify that the collected data remained complete throughout the extraction and storage process.\n- Direct download via FTP, SFTP, API or S3: Some datasets were acquired using secure transfer protocols such as FTP (File Transfer Protocol), SFTP (Secure File Transfer Protocol), or API (Application Programming Interface) requests from cloud storage services such as Amazon S3. As with the open direct download method, data integrity tests were used to validate the completeness of the files to ensure that the files were not altered or corrupted during the transfer process.\n\n**If the dataset is a sample from a larger set, what was the sampling strategy?**\n\nThe sampling strategy was to use the whole dataset resulting from the filtering explained in the 'preprocessing/cleaning/labelling' section, \nwith the particularity that an upsampling of 2 (i.e. twice the probability of sampling a document) was performed for the co-official languages \nof Spain (Spanish, Catalan, Galician, Basque), and a downsampling of 1/2 was applied for code (half the probability of sampling a code document, \nevenly distributed among all programming languages).\n\n**Who was involved in the data collection process and how were they compensated?**\n\nThis data is generally extracted, filtered and sampled by automated processes. The code required to run these processes has been developed entirely \nby members of the Language Technologies data team, or otherwise obtained from open-source software. Furthermore, there has been no monetary \nconsideration for acquiring data from suppliers.\n\n**Over what timeframe was the data collected? Does this timeframe match the creation timeframe of the data associated with the instances? If not, please describe the timeframe in which the data associated with the instances was created.**\n\nData were acquired and processed from April 2023 to April 2024. However, as mentioned, much data has been obtained from open projects such \nas Common Crawl, which contains data from 2014, so it is the end date (04/2024) rather than the start date that is important.\n\n**Were any ethical review processes conducted? If so, please provide a description of these review processes, including the outcomes, as well as a link or other access point to any supporting documentation.**\n\nNo particular ethical review process has been carried out as the data is mostly open and not particularly sensitive. However, we have an \ninternal evaluation team and a bias team to monitor ethical issues. In addition, we work closely with ‘Observatori d'Ètica en Intel·ligència \nArtificial’ (OEIAC) and ‘Agencia Española de Supervisión de la Inteligencia Artificial’ (AESIA) to audit the processes we carry out from an \nethical and legal point of view, respectively.\n\n#### Preprocessing\n\n**Was any preprocessing/cleaning/labeling of the data done? If so, please provide a description. If not, you may skip the remaining questions in this section.**\n\nNo changes were made to the content of individual text document instances. However, the web-sourced documents underwent a filtering process based on specific criteria along two key dimensions:\n- Quality filtering: The text processing pipeline CURATE (Palomar et. al, 2024) calculates a quality score for each document based on a set of filtering criteria that identify undesirable textual characteristics. Any document with a score below the 0.8 threshold was excluded from the dataset.\n- Harmful or adult content filtering: To reduce the amount of harmful or inappropriate material in the dataset, documents from Colossal OSCAR were filtered using the Ungoliant pipeline (Abadji et al., 2021), which uses the 'harmful\\_pp' field, a perplexity-based score generated by a language model.\n\n**Was the “raw” data saved in addition to the preprocessed/cleaned/labeled data? If so, please provide a link or other access point to the “raw” data.**\n\nThe original raw data was not kept.\n\n**Is the software that was used to preprocess/clean/label the data available? If so, please provide a link or other access point.**\n\nYes, the preprocessing and filtering software is open-sourced. The [CURATE](https://github.com/langtech-bsc/CURATE) pipeline was used for CATalog and other curated datasets, \nand the [Ungoliant](https://github.com/oscar-project/ungoliant) pipeline was used for the OSCAR project.\n\n#### Uses\n\n**Has the dataset been used for any tasks already? If so, please provide a description.**\n\nPre-train the Salamandra model family.\n\n**What (other) tasks could the dataset be used for?**\n\nThe data can be used primarily to pre-train other language models, which can then be used for a wide range of use cases. The dataset could \nalso be used for other tasks such as fine-tuning language models, cross-lingual NLP tasks, machine translation, domain-specific text \ngeneration, and language-specific data analysis.\n\n**Is there anything about the composition of the dataset or the way it was collected and preprocessed/cleaned/labeled that might impact future uses? Is there anything a dataset consumer could do to mitigate these risks or harms?**\n \nWeb-crawled content is over-represented with standard language varieties, impacting language model performance for minority languages. \nLanguage diversity in data is crucial to avoid bias, especially in encoding non-standard dialects, preventing the exclusion of demographic \ngroups. Moreover, despite legal uncertainties in web-scraped data, we prioritize permissive licenses and privacy protection measures, \nacknowledging the challenges posed by personally identifiable information (PII) within large-scale datasets. Our ongoing efforts aim to \naddress privacy concerns and contribute to a more inclusive linguistic dataset.\n\n**Are there tasks for which the dataset should not be used?**\n\n-\n\n#### Distribution \n\n**Will the dataset be distributed to third parties outside of the entity on behalf of which the dataset was created? If so, please provide a description.**\n\nThe dataset will not be released or distributed to third parties. Any related question to distribution is omitted in this section.\n\n#### Maintenance\n\n**Who will be supporting/hosting/maintaining the dataset?**\n\nThe dataset will be hosted by the Language Technologies unit (LangTech) of the Barcelona Supercomputing Center (BSC). The team will ensure \nregular updates and monitor the dataset for any issues related to content integrity, legal compliance, and bias for the sources they are \nresponsible for.\n\n**How can the owner/curator/manager of the dataset be contacted?**\n\nThe data owner may be contacted with the email address langtech@bsc.es.\n\n**Will the dataset be updated?**\n\nThe dataset will not be updated.\n\n**If the dataset relates to people, are there applicable limits on the retention of the data associated with the instances? If so, please describe these limits and explain how they will be enforced.**\n\nThe dataset does not keep sensitive data that could allow direct identification of individuals, apart from the data that is publicly available in \nweb-sourced content. Due to the sheer volume and diversity of web data, it is not feasible to notify individuals or manage data retention on an \nindividual basis. However, efforts are made to mitigate the risks associated with sensitive information through pre-processing and filtering to \nremove identifiable or harmful content. Despite these measures, vigilance is maintained to address potential privacy and ethical issues.\n\n**Will older versions of the dataset continue to be supported/hosted/maintained? If so, please describe how. If not, please describe how its obsolescence will be communicated to dataset consumers.**\n\nSince the dataset will not be updated, only the final version will be kept.\n\n**If others want to extend/augment/build on/contribute to the dataset, is there a mechanism for them to do so?**\n\nThe dataset does not allow for external contributions.\n\n</details>\n\n### Finetuning Data\n\nThis instructed-tuned variant has been fine-tuned with a collection of 273k instructions, focusing on the performance of Catalan, English and Spanish. However, instruction data for other closely related Iberian languages has also been included, since it yielded a positive impact on the languages of interest. That said, the performance in these additional languages is not guaranteed due to the limited amount of available data and the lack of resources for thorough testing.\n\n| **Dataset** | **ca** | **en** | **es** | **eu** | **gl** | **pt** | **Total** |\n|----------------------|------------|-------------|------------|-----------|---------|------------|-------------|\n| alpaca-cleaned | | 49,950 | | | | | **49,950** |\n| aya-dataset | | 3,941 | 3,851 | 939 | | 8,995 | **17,726** |\n| coqcat | 4,797 | | | | | | **4,797** |\n| databricks-dolly-15k | | 15,011 | | | | | **15,011** |\n| dolly-ca | 3,232 | | | | | | **3,232** |\n| flores-dev | 986 | 1,037 | 1,964 | 493 | 505 | | **4,985** |\n| mentor-ca | 7,119 | | | | | | **7,119** |\n| mentor-es | | | 7,122 | | | | **7,122** |\n| no-robots | | 9,485 | | | | | **9,485** |\n| oasst-ca | 2,517 | | | | | | **2,517** |\n| oasst2 | 750 | 31,086 | 15,438 | 190 | 197 | 1,203 | **48,864** |\n| open-orca | | 49,996 | | | | | **49,996** |\n| rag-multilingual | 16,043 | 14,997 | 11,263 | | | | **42,303** |\n| tower-blocks | | 7,762 | 1,000 | | | 1,000 | **9,762** |\n| **Total** | **35,444** | **183,265** | **40,638** | **1,622** | **702** | **11,198** | **272,869** |\n\n---\n\n\n## Evaluation\n\n### Gold-standard benchmarks\nWiP\n<!--\nEvaluation is done using the Language Model Evaluation Harness (Gao et al., 2024). We evaluate on a set of tasks taken from [SpanishBench](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/spanish_bench), [CatalanBench](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/catalan_bench), [BasqueBench](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/basque_bench) and [GalicianBench](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/galician_bench). These benchmarks include both new and existing tasks and datasets. Given that this is an instructed model, we add LM Evaluation Harness's native feature of `chat-template` to the setup. In the tables below, we include the results in a selection of evaluation datasets that represent model's performance across a variety of tasks within these benchmarks. \n\nWe only use tasks that are either human generated, human translated, or with a strong human-in-the-loop (i.e., machine translation followed by professional revision or machine generation followed by human revision and annotation). This is the reason behind the variety in number of tasks reported across languages. As more tasks that fulfill these requirements are published, we will update the presented results. We also intend to expand the evaluation to other languages, as long as the datasets meet our quality standards.\n\nDuring the implementation of the evaluation we observed a series of issues worth considering when replicating and interpreting the results presented. These issues include ≈1.5% variances in performance in some tasks depending on the version of the `transformers` library used, and depending on the use (or lack of use) of tensor parallelism when loading a model. When implementing existing tasks, we carry out a comprehensive quality evaluation of the dataset, the Harness task itself, and what kind of input models see during evaluation. Our implementation (see links above) addresses multiple existing problems such as errors in datasets and prompts, and lack of pre-processing. All this means that results will vary if using other Harness implementations, and may slightly vary depending on the replication setup.\n\nIt should be noted that these results are subject to all the drawbacks of every current gold-standard evaluation, and that the figures do not fully represent the model's capabilities and potential. We thus advise caution when reading and interpreting the results.\n\nA full list of results compared to other baselines, a discussion of the model's performance across tasks and its implications, and details regarding problem-solving with task implementation will soon be available in the technical report.\n\nAll results reported below are on a 0-shot setting.\n\n#### Spanish\n\n<table><thead>\n <tr>\n <th>Category</th>\n <th>Task</th>\n <th>Metric</th>\n <th>Result</th>\n </tr></thead>\n<tbody>\n <tr>\n <td>Commonsense Reasoning</td>\n <td>xstorycloze_es</td>\n <td>acc</td>\n <td>73.13</td>\n </tr>\n <tr>\n <td rowspan=\"2\">NLI</td>\n <td>wnli_es</td>\n <td>acc</td>\n <td>60.56</td>\n </tr>\n <tr>\n <td>xnli_es</td>\n <td>acc</td>\n <td>50.84</td>\n </tr>\n <tr>\n <td>Paraphrasing</td>\n <td>paws_es</td>\n <td>acc</td>\n <td>60.75</td>\n </tr>\n <tr>\n <td>QA</td>\n <td>xquad_es</td>\n <td>acc</td>\n <td>63.20/td>\n </tr>\n <tr>\n <td>Translation</td>\n <td>flores_es</td>\n <td>bleu</td>\n <td>14.95</td>\n </tr>\n</tbody>\n</table>\n\n#### Catalan\n\n<table><thead>\n <tr>\n <th>Category</th>\n <th>Task</th>\n <th>Metric</th>\n <th>Result</th>\n </tr></thead>\n<tbody>\n <tr>\n <td rowspan=\"2\">Commonsense Reasoning</td>\n <td>copa_ca</td>\n <td>acc</td>\n <td>82.80</td>\n </tr>\n <tr>\n <td>xstorycloze_ca</td>\n <td>acc</td>\n <td>73.73</td>\n </tr>\n <tr>\n <td rowspan=\"2\">NLI</td>\n <td>wnli_ca</td>\n <td>acc</td>\n <td>64.79</td>\n </tr>\n <tr>\n <td>xnli_ca</td>\n <td>acc</td>\n <td>53.45</td>\n </tr>\n <tr>\n <td rowspan=\"2\">Paraphrasing</td>\n <td>parafraseja</td>\n <td>acc</td>\n <td>64.15</td>\n </tr>\n <tr>\n <td>paws_ca</td>\n <td>acc</td>\n <td>64.35</td>\n </tr>\n <tr>\n <td rowspan=\"5\">QA</td>\n <td>arc_ca_easy</td>\n <td>acc</td>\n <td>73.57</td>\n </tr>\n <tr>\n <td>arc_ca_challenge</td>\n <td>acc</td>\n <td>45.90</td>\n </tr>\n <tr>\n <td>openbookqa_ca</td>\n <td>acc</td>\n <td>40.60</td>\n </tr>\n <tr>\n <td>piqa_ca</td>\n <td>acc</td>\n <td>73.39</td>\n </tr>\n <tr>\n <td>siqa_ca</td>\n <td>acc</td>\n <td>51.84</td>\n </tr>\n <tr>\n <td>Translation</td>\n <td>flores_ca</td>\n <td>bleu</td>\n <td>20.49</td>\n </tr>\n</tbody></table>\n\n#### Basque\n\n<table><thead>\n <tr>\n <th>Category</th>\n <th>Task</th>\n <th>Metric</th>\n <th>Result</th>\n </tr></thead>\n<tbody>\n <tr>\n <td rowspan=\"2\">Commonsense Reasoning</td>\n <td>xcopa_eu</td>\n <td>acc</td>\n <td>67.80</td>\n </tr>\n <tr>\n <td>xstorycloze_eu</td>\n <td>acc</td>\n <td>65.06</td>\n </tr>\n <tr>\n <td rowspan=\"2\">NLI</td>\n <td>wnli_eu</td>\n <td>acc</td>\n <td>56.34</td>\n </tr>\n <tr>\n <td>xnli_eu</td>\n <td>acc</td>\n <td>47.34</td>\n </tr>\n <tr>\n <td rowspan=\"3\">QA</td>\n <td>eus_exams</td>\n <td>acc</td>\n <td>45.98</td>\n </tr>\n <tr>\n <td>eus_proficiency</td>\n <td>acc</td>\n <td>43.92</td>\n </tr>\n <tr>\n <td>eus_trivia</td>\n <td>acc</td>\n <td>50.38</td>\n </tr>\n <tr>\n <td>Reading Comprehension</td>\n <td>eus_reading</td>\n <td>acc</td>\n <td>48.01</td>\n </tr>\n <tr>\n <td>Translation</td>\n <td>flores_eu</td>\n <td>bleu</td>\n <td>10.99</td>\n </tr>\n</tbody></table>\n\n#### Galician\n\n<table><thead>\n <tr>\n <th>Category</th>\n <th>Task</th>\n <th>Metric</th>\n <th>Result</th>\n </tr></thead>\n<tbody>\n <tr>\n <td rowspan=\"2\">Paraphrasing</td>\n <td>parafrases_gl</td>\n <td>acc</td>\n <td>58.50</td>\n </tr>\n <tr>\n <td>paws_gl</td>\n <td>acc</td>\n <td>62.45</td>\n </tr>\n <tr>\n <td>QA</td>\n <td>openbookqa_gl</td>\n <td>acc</td>\n <td>37.20</td>\n </tr>\n <tr>\n <td>Translation</td>\n <td>flores_gl</td>\n <td>bleu</td>\n <td>18.81</td>\n </tr>\n</tbody>\n</table>\n-->\n\n### LLM-as-a-judge\n\nWe use [Prometheus-2 8x7B](https://huggingface.co/prometheus-eval/prometheus-8x7b-v2.0) as a judge to evaluate the responses of the model. Tasks are created from existing multilingual evaluation datasets covering the same categories as the ones measured in our gold-standard benchmarks. We randomly select a subset of 250 instances per language from the `test` set of each source dataset. To evaluate the responses of our model, we use task-specific criteria developed in-house for the _LLM-judge_ to use. Each criterion is measured either as a 5-point Likert scale or as a binary task depending on the idiosyncrasy of the task and criterion.\n\nPrompts for each task are created in various ways to score the model's robustness in addition to these criteria. This is done by presenting the same source instance within three different prompts. We then calculate the variance between the scores assigned by the _LLM-judge_ to our model's responses to the three prompt styles and average it across all instances. Prompts are human translated to all languages measured. We do not provide the _LLM-judge_ with a reference answer.\n\nThe _judge_ prompt we use during evaluation is the same used to fine tune the Prometheus-2 family. We keep the _judge_ prompt and criteria used to present the _LLM-judge_ with the task prompts and model responses in English for evaluation across languages. The _judge_ prompt used is:\n\n```python\n\"You are a fair judge assistant tasked with providing clear, objective feedback based on specific criteria, ensuring each assessment reflects the absolute standards set for performance.\n\n###Task Description:\nAn instruction (might include an Input inside it), a response to evaluate, and a score rubric representing a evaluation criteria are given.\n1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.\n2. After writing a feedback, write a score that is an integer between {a} and {b}. You should refer to the score rubric.\n3. The output format should look as follows: \\\"Feedback: (write a feedback for criteria) [RESULT] (an integer number between {a} and {b})\\\"\n4. Please do not generate any other opening, closing, and explanations.\n\n###The instruction to evaluate:\n{input}\n\n###Response to evaluate:\n{prediction}\n\n###Score Rubrics:\n{criteria}\n\n###Feedback:\"\n```\n\nAs an example, prompts for the Math task in English are based on instances from [MGSM](https://huggingface.co/datasets/juletxara/mgsm), and each instance is presented within these prompts:\n\n```python\n\"en\": [\n (\"I need help with this math problem: \\\"\", \"\\\" Give me the answer step by step and also the final result separately.\"),\n (\"Can you please help me answer this? \\\"\", \"\\\" Explain the answer and give me the final result as well. Thanks.\"),\n (\"Help me with this problem: \\\"\", \"\\\" I need the answer explained and the final result separately.\")\n]\n```\n\n\nThis task is then evaluated by the _LLM-judge_ using two criteria, reasoning capability (5-point Likert) and mathematical correctness (binary):\n\n```python\nreasoning_capability_criteria = {\n \"reasoning_capability\": \"\"\"\n[Does the model's answer demonstrate reasoning capability?]\nScore 1: The answer demonstrates poor reasoning, with illogical arguments or conclusions that do not follow from the provided information.\nScore 2: The answer shows weak reasoning, with some logical connections but also contains significant flaws or gaps in the argumentation.\nScore 3: The answer demonstrates adequate reasoning, with generally logical arguments, but may have minor flaws or a lack of depth in the reasoning process.\nScore 4: The answer shows strong reasoning, with well-structured arguments and conclusions that logically follow from the information provided.\nScore 5: The answer demonstrates exceptional reasoning, with clear, coherent, and insightful arguments that are logically sound and well-supported by the information provided.\"\"\"\n}\n\nmathematical_correctness_binary_criteria = {\n \"mathematical_correctness_binary\": \"\"\"\n[Is the model's answer mathematically correct?]\nScore 0: The answer contains mathematical errors that render the solution incorrect or unreliable.\nScore 1: The answer is mathematically correct, with accurate calculations and appropriate use of mathematical concepts.\"\"\"\n}\n```\n\n#### Multilingual results\n\nHere, we present results for seven categories of tasks in Spanish, Catalan, Basque, Galician, and English. Results are presented for each task, criterion and language. Criteria with a `(B)` after their name are binary criteria (i.e., numbers go from 0 to 1, where 1 is best). The rest of the criteria are measured using a 5-point Likert scale, where 5 is best. The first number of the pair of numbers separated by `/` shows the average score for the criterion (and language). The second number of each pair is the robustness score, where numbers closer to 0 mean that the model generates similar responses when comparing the three prompt varieties for a single instance.\n\nFurther details on all tasks and criteria, a full list of results compared to other baselines, a discussion of the model's performance across tasks and its implications, and details regarding problem-solving with task implementation will soon be available in the technical report.\n\n<table class=\"tg\"><thead>\n <tr>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">Category</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">Dataset</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">Criteria</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">es</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">ca</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">gl</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">eu</span></th>\n <th class=\"tg-0pky\"><span style=\"font-weight:bold\">en</span></th>\n </tr></thead>\n<tbody>\n <tr>\n <td class=\"tg-0pky\">Commonsense Reasoning</td>\n <td class=\"tg-0pky\">XStoryCloze</td>\n <td class=\"tg-0pky\">Ending coherence</td>\n <td class=\"tg-0pky\">3.24/0.63</td>\n <td class=\"tg-0pky\">3.12/0.51</td>\n <td class=\"tg-0pky\">2.87/0.59</td>\n <td class=\"tg-0pky\">2.16/0.52</td>\n <td class=\"tg-0pky\">3.71/0.50</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\" rowspan=\"3\">Paraphrasing</td>\n <td class=\"tg-0pky\" rowspan=\"3\">PAWS</td>\n <td class=\"tg-0pky\">Completeness `(B)`</td>\n <td class=\"tg-0pky\">0.86/0.07</td>\n <td class=\"tg-0pky\">0.82/0.09</td>\n <td class=\"tg-0pky\">0.78/0.10</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">0.92/0.05</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Paraphrase generation</td>\n <td class=\"tg-0pky\">3.81/0.54</td>\n <td class=\"tg-0pky\">3.67/0.55</td>\n <td class=\"tg-0pky\">3.56/0.57</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">3.98/0.37</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Grammatical correctness `(B)`</td>\n <td class=\"tg-0pky\">0.93/0.03</td>\n <td class=\"tg-0pky\">0.92/0.05</td>\n <td class=\"tg-0pky\">0.89/0.06</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">0.96/0.03</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\" rowspan=\"2\">Reading Comprehension</td>\n <td class=\"tg-0pky\" rowspan=\"2\">Belebele</td>\n <td class=\"tg-0pky\">Passage comprehension</td>\n <td class=\"tg-0pky\">3.43/0.43</td>\n <td class=\"tg-0pky\">3.28/0.50</td>\n <td class=\"tg-0pky\">3.02/0.56</td>\n <td class=\"tg-0pky\">2.61/0.43</td>\n <td class=\"tg-0pky\">3.43/0.58</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Answer relevance `(B)`</td>\n <td class=\"tg-0pky\">0.86/0.05</td>\n <td class=\"tg-0pky\">0.84/0.05</td>\n <td class=\"tg-0pky\">0.75/0.08</td>\n <td class=\"tg-0pky\">0.65/0.11</td>\n <td class=\"tg-0pky\">0.83/0.06</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\" rowspan=\"2\">Extreme Summarization</td>\n <td class=\"tg-0pky\" rowspan=\"2\">XLSum & caBreu & summarization_gl</td>\n <td class=\"tg-0pky\">Informativeness</td>\n <td class=\"tg-0pky\">3.37/0.34</td>\n <td class=\"tg-0pky\">3.57/0.31</td>\n <td class=\"tg-0pky\">3.40/0.31</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">3.32/0.26</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Conciseness</td>\n <td class=\"tg-0pky\">3.06/0.34</td>\n <td class=\"tg-0pky\">2.88/0.50</td>\n <td class=\"tg-0pky\">3.09/0.38</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">3.32/0.22</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\" rowspan=\"2\">Math</td>\n <td class=\"tg-0pky\" rowspan=\"2\">MGSM</td>\n <td class=\"tg-0pky\">Reasoning capability</td>\n <td class=\"tg-0pky\">3.29/0.72</td>\n <td class=\"tg-0pky\">3.16/0.65</td>\n <td class=\"tg-0pky\">3.33/0.60</td>\n <td class=\"tg-0pky\">2.56/0.52</td>\n <td class=\"tg-0pky\">3.35/0.65</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Mathematical correctness `(B)`</td>\n <td class=\"tg-0pky\">0.68/0.12</td>\n <td class=\"tg-0pky\">0.65/0.13</td>\n <td class=\"tg-0pky\">0.73/0.11</td>\n <td class=\"tg-0pky\">0.59/0.13</td>\n <td class=\"tg-0pky\">0.67/0.12</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\" rowspan=\"2\">Translation form Language</td>\n <td class=\"tg-0pky\" rowspan=\"2\">FLORES-200</td>\n <td class=\"tg-0pky\">Fluency</td>\n <td class=\"tg-0pky\">3.95/0.11</td>\n <td class=\"tg-0pky\">3.88/0.15</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">3.92/0.14</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Accuracy</td>\n <td class=\"tg-0pky\">4.22/0.15</td>\n <td class=\"tg-0pky\">4.25/0.21</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">4.25/0.23</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\" rowspan=\"2\">Translation to Language</td>\n <td class=\"tg-0pky\" rowspan=\"2\">FLORES-200</td>\n <td class=\"tg-0pky\">Fluency</td>\n <td class=\"tg-0pky\">3.92/0.11</td>\n <td class=\"tg-0pky\">3.84/0.14</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">4.19/0.14</td>\n </tr>\n <tr>\n <td class=\"tg-0pky\">Accuracy</td>\n <td class=\"tg-0pky\">4.31/0.16</td>\n <td class=\"tg-0pky\">4.18/0.20</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">-- / --</td>\n <td class=\"tg-0pky\">4.63/0.15</td>\n </tr>\n</tbody></table>\n\n---\n\n## Ethical Considerations and Limitations\n\nWe examine the presence of undesired societal and cognitive biases present in this model using different benchmarks. For societal biases, \nwe test performance using the BBQ dataset (Parrish et al., 2022) in the original English and the Regard dataset (Sheng et al., 2019). \nWe report that while performance is high (accuracies around 0.8 depending on the social category) in disambiguated settings, \nthe model performs very poorly in ambiguous settings, which indicates the presence of societal biases that need to be further addressed in post-training phases.\n\nOur cognitive bias analysis focuses on positional effects in 0-shot settings, and majority class bias in few-shot settings. \nFor positional effects, we leverage the ARC Multiple Choice Question dataset (Clark et al., 2018). We observe significant, \nbut relatively weak primacy effects, whereby the model shows a preference for answers towards the beginning of the list of provided answers. \nWe measure the effects of majority class effects in few-shot settings using SST-2 (Socher et al., 2013). We again detect significant effects, \nwith a small effect size. This suggests that the model is relatively robust against the examined cognitive biases.\n\n We highlight that our analyses of these biases are by no means exhaustive and are limited by the relative scarcity of adequate resources \n in all languages present in the training data. We aim to gradually extend and expand our analyses in future work.\n \n These results can be expected from a model that has undergone only a preliminary instruction tuning. \n These tests are performed in order to show the biases the model may contain. We urge developers to take \n them into account and perform safety testing and tuning tailored to their specific applications of the model.\n\n---\n\n## Additional information\n\n### Author\nThe Language Technologies Unit from Barcelona Supercomputing Center.\n\n### Contact\nFor further information, please send an email to <langtech@bsc.es>.\n\n### Copyright\nCopyright(c) 2024 by Language Technologies Unit, Barcelona Supercomputing Center.\n\n### Funding\nThis work has been promoted and financed by the Government of Catalonia through the [Aina Project](https://projecteaina.cat/).\n\nThis work is funded by the _Ministerio para la Transformación Digital y de la Función Pública_ - Funded by EU – NextGenerationEU \nwithin the framework of [ILENIA Project](https://proyectoilenia.es/) with reference 2022/TL22/00215337.\n\n### Acknowledgements\n\nThis project has benefited from the contributions of numerous teams and institutions, mainly through data contributions, knowledge transfer or technical support. \n\nIn Catalonia, many institutions have been involved in the project. Our thanks to Òmnium Cultural, Parlament de Catalunya, Institut d'Estudis Aranesos, Racó Català, Vilaweb, ACN, Nació Digital, El món and Aquí Berguedà.\n\nAt the national level, we are especially grateful to our ILENIA project partners: CENID, HiTZ and CiTIUS for their participation. We also extend our genuine gratitude to the Spanish Senate and Congress, Fundación Dialnet, and the ‘Instituto Universitario de Sistemas Inteligentes y Aplicaciones Numéricas en Ingeniería (SIANI)’ of the University of Las Palmas de Gran Canaria. \n\nAt the international level, we thank the Welsh government, DFKI, Occiglot project, especially Malte Ostendorff, and The Common Crawl Foundation, especially Pedro Ortiz, for their collaboration. We would also like to give special thanks to the NVIDIA team, with whom we have met regularly, specially to: Ignacio Sarasua, Adam Henryk Grzywaczewski, Oleg Sudakov, Sergio Perez, Miguel Martinez, Felipes Soares and Meriem Bendris. Their constant support has been especially appreciated throughout the entire process.\n\nTheir valuable efforts have been instrumental in the development of this work.\n\n### Disclaimer\nBe aware that the model may contain biases or other unintended distortions. \nWhen third parties deploy systems or provide services based on this model, or use the model themselves, \nthey bear the responsibility for mitigating any associated risks and ensuring compliance with applicable regulations, \nincluding those governing the use of Artificial Intelligence.\n\nThe Barcelona Supercomputing Center, as the owner and creator of the model, shall not be held liable for any outcomes resulting from third-party use.\n\n### Citation\n```\n@misc{gonzalezagirre2025salamandratechnicalreport,\n title={Salamandra Technical Report}, \n author={Aitor Gonzalez-Agirre and Marc Pàmies and Joan Llop and Irene Baucells and Severino Da Dalt and Daniel Tamayo and José Javier Saiz and Ferran Espuña and Jaume Prats and Javier Aula-Blasco and Mario Mina and Adrián Rubio and Alexander Shvets and Anna Sallés and Iñaki Lacunza and Iñigo Pikabea and Jorge Palomar and Júlia Falcão and Lucía Tormo and Luis Vasquez-Reina and Montserrat Marimon and Valle Ruíz-Fernández and Marta Villegas},\n year={2025},\n eprint={2502.08489},\n archivePrefix={arXiv},\n primaryClass={cs.CL},\n url={https://arxiv.org/abs/2502.08489}, \n}\n```\n\n### License\n[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)\n\n## Model Index\n|Model|Base|Instruct|\n|:---:|:---:|:---:|\n|2B| [Link](https://huggingface.co/BSC-LT/salamandra-2b) | [Link](https://huggingface.co/BSC-LT/salamandra-2b-instruct) |\n|7B| [Link](https://huggingface.co/BSC-LT/salamandra-7b) | [Link](https://huggingface.co/BSC-LT/salamandra-7b-instruct) |\n|40B| [Link](https://huggingface.co/BSC-LT/ALIA-40b) | WiP |\n"},"matched_task":{"kind":"list like","value":["QUESTION_ANSWERING","TRANSLATION","SUMMARIZATION","PARAPHRASING"],"string":"[\n \"QUESTION_ANSWERING\",\n \"TRANSLATION\",\n \"SUMMARIZATION\",\n \"PARAPHRASING\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["BEAR","SCIELO"],"string":"[\n \"BEAR\",\n \"SCIELO\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"Non_BioNLP"}}},{"rowIdx":297,"cells":{"id":{"kind":"string","value":"BSC-NLP4BIA/bsc-bio-ehr-es-meddocan"},"author":{"kind":"string","value":"BSC-NLP4BIA"},"task_category":{"kind":"string","value":"token-classification"},"tags":{"kind":"list like","value":["transformers","pytorch","roberta","token-classification","biomedical","clinical","EHR","spanish","anonymization","es","base_model:PlanTL-GOB-ES/bsc-bio-ehr-es","base_model:finetune:PlanTL-GOB-ES/bsc-bio-ehr-es","license:apache-2.0","model-index","autotrain_compatible","endpoints_compatible","region:us"],"string":"[\n \"transformers\",\n \"pytorch\",\n \"roberta\",\n \"token-classification\",\n \"biomedical\",\n \"clinical\",\n \"EHR\",\n \"spanish\",\n \"anonymization\",\n \"es\",\n \"base_model:PlanTL-GOB-ES/bsc-bio-ehr-es\",\n \"base_model:finetune:PlanTL-GOB-ES/bsc-bio-ehr-es\",\n \"license:apache-2.0\",\n \"model-index\",\n \"autotrain_compatible\",\n \"endpoints_compatible\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1717,"string":"1,717"},"last_modified":{"kind":"number","value":1722,"string":"1,722"},"downloads":{"kind":"number","value":15,"string":"15"},"likes":{"kind":"number","value":0,"string":"0"},"README":{"kind":"string","value":"---\nbase_model:\n- PlanTL-GOB-ES/bsc-bio-ehr-es\nlanguage:\n- es\nlicense: apache-2.0\nmetrics:\n- precision\n- recall\n- f1\ntags:\n- biomedical\n- clinical\n- EHR\n- spanish\n- anonymization\nwidget:\n- text: El diagnóstico definitivo de nuestro paciente fue de un Adenocarcinoma de\n pulmón cT2a cN3 cM1a Estadio IV (por una única lesión pulmonar contralateral)\n PD-L1 90%, EGFR negativo, ALK negativo y ROS-1 negativo.\n- text: Durante el ingreso se realiza una TC, observándose un nódulo pulmonar en el\n LII y una masa renal derecha indeterminada. Se realiza punción biopsia del nódulo\n pulmonar, con hallazgos altamente sospechosos de carcinoma.\n- text: Trombosis paraneoplásica con sospecha de hepatocarcinoma por imagen, sobre\n hígado cirrótico, en paciente con índice Child-Pugh B.\nmodel-index:\n- name: BSC-NLP4BIA/bsc-bio-ehr-es-meddocan\n results:\n - task:\n type: token-classification\n dataset:\n name: MEDDOCAN\n type: MEDDOCAN\n metrics:\n - type: precision\n value: 0.95\n name: precision (micro)\n - type: recall\n value: 0.972\n name: recall (micro)\n - type: f1\n value: 0.961\n name: f1 (micro)\n - task:\n type: token-classification\n dataset:\n name: carmen-anonymization\n type: carmen-anonymization\n metrics:\n - type: precision\n value: 0.804\n name: precision (micro)\n - type: recall\n value: 0.818\n name: recall (micro)\n - type: f1\n value: 0.811\n name: f1 (micro)\n---\n\n\n# Spanish RoBERTa-base biomedical model finetuned for the Named Entity Recognition (NER) task on the MEDDOCAN dataset.\n\n## Table of contents\n<details>\n<summary>Click to expand</summary>\n\n- [Model description](#model-description)\n- [Intended uses and limitations](#intended-use)\n- [How to use](#how-to-use)\n- [Limitations and bias](#limitations-and-bias)\n- [Training](#training)\n- [Evaluation](#evaluation)\n- [Additional information](#additional-information)\n - [Authors](#authors)\n - [Contact information](#contact-information)\n - [Licensing information](#licensing-information)\n - [Funding](#funding)\n - [Citing information](#citing-information)\n - [Disclaimer](#disclaimer)\n \n</details>\n\n## Model description\nA fine-tuned version of the [bsc-bio-ehr-es](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es) model on the [MEDDOCAN](https://zenodo.org/records/4279323) corpus.\n\nFor further information, check the [official website](https://temu.bsc.es/meddocan/).\n\n## Intended uses and limitations\n\nTDB\n\n## How to use\n\n⚠ We recommend pre-tokenizing the input text into words instead of providing it directly to the model, as this is how the model was trained. Otherwise, the results and performance might get affected.\n\nA usage example can be found [here](https://github.com/nlp4bia-bsc/hugging-face-pipeline/blob/main/simple_inference.ipynb).\n\n## Limitations and bias\nAt the time of submission, no measures have been taken to estimate the bias embedded in the model. However, we are well aware that our models may be biased since the corpora have been collected using crawling techniques on multiple web sources. We intend to conduct research in these areas in the future, and if completed, this model card will be updated. \n\n## Training\nThe model was trained using the Barcelona Supercomputing Center infrastructure.\n\n## Evaluation \nF1 Score: 0.961 on MEDDOCAN\n\nF1 Score: 0.811 on CARMEN-I Anonymization\n\n## Additional information\n\n### Authors\nNLP4BIA team at the Barcelona Supercomputing Center (nlp4bia@bsc.es).\n\n### Contact information\njan.rodriguez [at] bsc.es\n\n### Licensing information\n[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)\n\n### Funding\nThis project was partially funded by the Spanish Plan for the Advancement of Language Technology (Plan TL) in collaboration with the Barcelona Supercomputing Center (BSC) and the Hospital Clinic de Barcelona (HCB). On the BSC's side, we acknowledge additional funding by the Spanish National AI4ProfHealth project (PID2020-119266RA-I00 MICIU/AEI/10.13039/501100011033) and EU Horizon projects (AI4HF 101080430 and DataTools4Heart 101057849). On the HCB's side, the project was also supported by the Instituto de Salud Carlos III (ISCIII).\n\n### Citing information\nPlease cite the following works:\n\n```\n@inproceedings{meddocan,\n title={{Automatic De-identification of Medical Texts in Spanish: the MEDDOCAN Track, Corpus, Guidelines, Methods and Evaluation of Results}},\n author={Marimon, Montserrat and Gonzalez-Agirre, Aitor and Intxaurrondo, Ander and Villegas, Marta and Krallinger, Martin},\n booktitle=\"Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2019)\",\n year={2019}\n}\n\n@misc{carmen_physionet, \n author = {Farre Maduell, Eulalia and Lima-Lopez, Salvador and Frid, Santiago Andres and Conesa, Artur and Asensio, Elisa and Lopez-Rueda, Antonio and Arino, Helena and Calvo, Elena and Bertran, Maria Jesús and Marcos, Maria Angeles and Nofre Maiz, Montserrat and Tañá Velasco, Laura and Marti, Antonia and Farreres, Ricardo and Pastor, Xavier and Borrat Frigola, Xavier and Krallinger, Martin}, \n title = {{CARMEN-I: A resource of anonymized electronic health records in Spanish and Catalan for training and testing NLP tools (version 1.0.1)}}, \n year = {2024}, \n publisher = {PhysioNet}, \n url = {https://doi.org/10.13026/x7ed-9r91} \n}\n\n@article{physionet,\n author = {Ary L. Goldberger and Luis A. N. Amaral and Leon Glass and Jeffrey M. Hausdorff and Plamen Ch. Ivanov and Roger G. Mark and Joseph E. Mietus and George B. Moody and Chung-Kang Peng and H. Eugene Stanley },\n title = {PhysioBank, PhysioToolkit, and PhysioNet },\n journal = {Circulation},\n volume = {101},\n number = {23},\n pages = {e215-e220},\n year = {2000},\n doi = {10.1161/01.CIR.101.23.e215},\n URL = {https://www.ahajournals.org/doi/abs/10.1161/01.CIR.101.23.e215}\n}\n```\n\n### Disclaimer\n\nThe models published in this repository are intended for a generalist purpose and are available to third parties. These models may have bias and/or any other undesirable distortions.\n\nWhen third parties, deploy or provide systems and/or services to other parties using any of these models (or using systems based on these models) or become users of the models, they should note that it is their responsibility to mitigate the risks arising from their use and, in any event, to comply with applicable regulations, including regulations regarding the use of artificial intelligence.\n\n---\n\nLos modelos publicados en este repositorio tienen una finalidad generalista y están a disposición de terceros. Estos modelos pueden tener sesgos y/u otro tipo de distorsiones indeseables.\n\nCuando terceros desplieguen o proporcionen sistemas y/o servicios a otras partes usando alguno de estos modelos (o utilizando sistemas basados en estos modelos) o se conviertan en usuarios de los modelos, deben tener en cuenta que es su responsabilidad mitigar los riesgos derivados de su uso y, en todo caso, cumplir con la normativa aplicable, incluyendo la normativa en materia de uso de inteligencia artificial."},"matched_task":{"kind":"list like","value":["NAMED_ENTITY_RECOGNITION"],"string":"[\n \"NAMED_ENTITY_RECOGNITION\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["MEDDOCAN"],"string":"[\n \"MEDDOCAN\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"BioNLP"}}},{"rowIdx":298,"cells":{"id":{"kind":"string","value":"samchain/econo-sentence-v2"},"author":{"kind":"string","value":"samchain"},"task_category":{"kind":"string","value":"sentence-similarity"},"tags":{"kind":"list like","value":["sentence-transformers","safetensors","bert","sentence-similarity","feature-extraction","generated_from_trainer","dataset_size:45044","loss:CoSENTLoss","economics","finance","en","dataset:samchain/econo-pairs-v2","arxiv:1908.10084","base_model:samchain/EconoBert","base_model:finetune:samchain/EconoBert","license:apache-2.0","model-index","autotrain_compatible","text-embeddings-inference","endpoints_compatible","region:us"],"string":"[\n \"sentence-transformers\",\n \"safetensors\",\n \"bert\",\n \"sentence-similarity\",\n \"feature-extraction\",\n \"generated_from_trainer\",\n \"dataset_size:45044\",\n \"loss:CoSENTLoss\",\n \"economics\",\n \"finance\",\n \"en\",\n \"dataset:samchain/econo-pairs-v2\",\n \"arxiv:1908.10084\",\n \"base_model:samchain/EconoBert\",\n \"base_model:finetune:samchain/EconoBert\",\n \"license:apache-2.0\",\n \"model-index\",\n \"autotrain_compatible\",\n \"text-embeddings-inference\",\n \"endpoints_compatible\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1739,"string":"1,739"},"last_modified":{"kind":"number","value":1739,"string":"1,739"},"downloads":{"kind":"number","value":64,"string":"64"},"likes":{"kind":"number","value":2,"string":"2"},"README":{"kind":"string","value":"---\nbase_model: samchain/EconoBert\ndatasets:\n- samchain/econo-pairs-v2\nlanguage:\n- en\nlibrary_name: sentence-transformers\nlicense: apache-2.0\nmetrics:\n- pearson_cosine\n- spearman_cosine\npipeline_tag: sentence-similarity\ntags:\n- sentence-transformers\n- sentence-similarity\n- feature-extraction\n- generated_from_trainer\n- dataset_size:45044\n- loss:CoSENTLoss\n- economics\n- finance\nwidget:\n- source_sentence: a consumer protection point of view, including through remuneration\n arrangements. failure to manage conduct risks can expose a financial institution\n to a variety of other risks which, if not managed properly, can threaten its solvency\n and sustainability. the regulatory regime for market conduct therefore provides\n a framework for the identification and management of conduct risk as a complementary\n framework to prudential regulation. this is part of the motivation for rigorous\n coordination and cooperation arrangements between the pa and the fsca envisaged\n by the financial sector regulation bill aimed at ensuring that all risks are holistically\n managed within an overarching financial stability policy framework. strengthening\n conduct in wholesale otc markets another initiative has been the development of\n a code of conduct for south african wholesale over - the - counter ( otc ) financial\n markets in a collaborative effort between regulators and key market participants.\n i will return to this aspect later. against the background of various investigations\n undertaken in many foreign jurisdictions in relation to foreign exchange market\n manipulation, the sarb and the fsb launched a review bis central bankers ’ speeches\n of the foreign exchange trading operations of south african authorised dealers\n in october 2014. as the rand is a globally traded currency, the aim of the review\n was to establish whether there may have been any spillover into our markets in\n relation to any misconduct or malpractice. it is important to note that, unlike\n in other jurisdictions, the south african review was not informed by whistle -\n blowing or any allegations – or indeed concrete evidence – of any misconduct.\n we had no evidence of widespread malpractice in the south african foreign exchange\n market but felt that, given the broad - based nature of investigations in other\n jurisdictions ( which also involved trading in emerging market currencies ), it\n would be prudent to obtain comfort that our foreign exchange trading practices\n were in line with best practice. it was therefore a proactive step on the part\n of south african regulators. the foreign exchange review committee established\n for this purpose – chaired by former senior deputy governor james cross, who is\n with us today – released its report in october 2015. the committee reported that\n it had found no evidence of manipulation or serious misconduct in the domestic\n foreign exchange market during the period covered by the review, but that there\n was scope for improvement in relation to governance and conduct. the committee\n also recommended that legislation be enhanced to give market conduct regulators\n wider powers to strengthen enforcement. south african regulators are in conversation\n with each other on how best to give effect to the implementation of the recommendations.\n there was also the recommendation\n sentences:\n - 'luigi federico signorini : g20 sustainable finance working group private sector\n roundtable welcome address by mr luigi federico signorini, senior deputy governor\n of the bank of italy, at the g20 sustainable finance working group private sector\n roundtable, online event, 17 may 2021. * * * welcome, and a good day to you all.\n i am happy to open the private sector roundtable, an event promoted by the g20\n presidency and by the chinese and american co - chairs of the sustainable finance\n working group. the roundtable will focus on the role of finance in helping fight\n climate change and promoting lowcarbon transition. the g20 finance ministers and\n central bank governors recently recognised the need to ‘ shape the current economic\n recovery by investing in innovative technologies and promoting just transitions\n toward more sustainable economies and societies ’. low - carbon transition is\n urgent and must be accelerated : the later we act, the greater the costs. it requires\n an unprecedented and unremitting effort. while quantitative estimates vary, the\n investments needed for transition are certainly huge ; they need to be sustained\n for a long time. governments have a central role in that they need to point the\n way by adopting an appropriate policy framework. a clear and credible path for\n government regulatory and fiscal action is also a prerequisite for efficient choices\n on the part of private finance. indeed, while many governments will directly invest\n their own money in many countries and mdbs will play their part, it is likely\n that the private sector will be called upon to finance most transition investment.\n there will be no transition without a general awareness of the need for it and\n a willingness, even a desire, to finance it. there are in fact quite a few encouraging\n signs. since last year, we have seen an explosion of ‘ net - zero commitments\n ’ in the private sector — though such commitments ( i am told ) are still confined\n to one sixth of publicly listed companies globally. at the same time, the appetite\n of ultimate investors and asset managers for ‘ green ’ investment is growing fast.\n i am sure many in the audience will have a clear perception of this fact. however,\n the path is still fraught with difficulties. on the market side, while sustainable\n finance is increasingly popular, it suffers from a lack of clear definitions and\n standards. ‘ greenwashing ’ is a danger ; good data, an agreed taxonomy, and adequate\n company disclosure are necessary. global consistency is important, as fragmentation\n of standards across jurisdictions is confusing for investors and costly for companies.\n standards are currently being drafted'\n - later, by more. so we raised interest rates through the second half of last year\n - and again in june - trying, as best we could through our tactics, to minimise\n any further unwanted upward pressure on sterling. but things have now clearly\n moved on. the outlook for the world economy deteriorated further through the summer\n under the impact of a series of new shocks. japan, the world ’ s second largest\n economy, slipped further into recession. russia - which had only weeks earlier\n embarked on an imf program - saw the collapse of the rouble and default on its\n debt. and acute nervousness spread through many of the world ’ s financial markets.\n although there has been some improvement in sentiment over the past month or two,\n and although the us and european economies continue to expand, the likelihood\n remains that world economic growth will be significantly slower than had been\n expected earlier in the summer. slower growth of world activity is bound to prolong\n the restraining external effect on growth and inflation in the uk, even though\n the exchange rate has now started to weaken. at the same time there are also now\n clearer signs of overall slowdown in our own economy. the evidence for this is\n less obvious in the backwards - looking economic and monetary data than it is\n in the forward - looking surveys, but even so the data suggest that we are beginning\n to see an easing of pressure, including an easing of pressure in the labour market.\n and the surveys themselves now point to a slowdown in service sector growth, including\n retail distribution, as well as a sharper decline in manufacturing output. this\n prospect is consistent with the reports which we receive directly from the bank\n ’ s network of regional agents and their 7000 - odd industrial and commercial\n contacts around the country. of course we pay very careful attention to this forward\n - looking evidence of developments in the economy alongside the data, and, like\n others, we have revised down our forecasts for output growth and inflation. and\n we have eased monetary policy quite sharply in the past two months, in the light\n of that evidence. our current best guess - published in last week ’ s inflation\n report is that, after the interest rate cuts, the growth of overall output next\n year will be around 1 %, picking up through the millennium to around trend in\n the second half of the year 2000. meanwhile, we expect underlying inflation to\n remain close to the target rate of 21 / 2 % - though perhaps a little above that\n rate during the course of next year. now no - one likes to see the\n - 'daniel mminele : conduct and culture in the banking and financial sectors opening\n address by mr daniel mminele, deputy governor of the south african reserve bank,\n at the g - 30 forum on banking conduct and culture, pretoria, 18 february 2016.\n * * * governor kganyago ( sarb ), governor sithole ( central bank of swaziland\n ), deputy governor mlambo ( reserve bank of zimbabwe ), deputy governors, groepe\n and naidoo ( sarb ), second deputy governor sullivan ( central bank of seychelles\n ), sir david walker ( vice chair of the group of thirty steering committee ),\n dr stuart mackintosh ( executive director of the g30 ), ms maria ramos ( chief\n executive officer of barclays africa ), the leadership of banks and other financial\n institutions, panel members, and esteemed delegates. it is a privilege and an\n honour for me to welcome you, on behalf of south african reserve bank ( sarb ),\n to this forum on banking conduct and culture, which we are co - hosting with the\n g30 and barclays africa. the g - 30 has, over the years, played a significant\n role in bringing together members of the banking, financial and regulatory community\n to discuss issues of common concern and examine the choices available to market\n practitioners and policymakers. given the enormous trust deficit that has built\n up since the global financial crisis, the topic of conduct and culture is of great\n importance and highly relevant to the global banking and financial sector. bankers\n have always had a delicate relationship with the societies they serve. it would\n appear that, at any point in time, it is almost a national sport across the globe\n to take a swipe at bankers. mark twain famously said : “ a banker is a fellow\n who lends you his umbrella when the sun is shining and wants it back the minute\n it begins to rain. ” and j m keynes asked : “ how long will it take to pay city\n men so entirely out of proportion to what other servants of society commonly received\n for performing social services not less useful or difficult? ” we would be terribly\n misguided to treat the current wave of discontent and deep mistrust as just another\n wave that will eventually subside. the most recent global financial crisis, from\n which almost nine years later we are still struggling to recover, shook the very\n foundations of our financial system and almost caused its total meltdown. in a\n nutshell, the crisis was about failures in conduct, culture, and supervisory practices.\n the consequences'\n- source_sentence: of our central bank distribution scheme have grown in number to\n embrace the physical quality of notes in circulation and the denominational mix.\n the last of these – denominational mix – poses the biggest challenge, but in the\n uk i believe we now have evidence to support the business case for atm and retailer\n dispense of £5s. and, finally, i am sometimes asked why this matters to the bank\n of england? after all, i can assure you that in the current financial conditions\n we are not short of difficult challenges. the answer is simple. we should not\n forget that our job is to ensure that confidence in our currency is maintained\n – and crucial to that is satisfying the public ’ s demand for our notes.\n sentences:\n - '##factory \". but the two clouds are moving towards us. so let us examine them.\n the outlook for the world economy has deteriorated markedly in recent months as\n a result of the sudden slowdown in the united states and signs of renewed stagnation\n in japan. the speed of the deterioration was a surprise, but not the fact of a\n slowdown. last year saw the fastest growth rate of the world economy for twelve\n years. growth in the us reached an annual rate of 6 % in the second quarter of\n last year, well above even optimistic estimates of sustainable growth rates. a\n slowdown was not only inevitable ; it was desirable. the main surprise in the\n us was the sharp and sudden break in both business and consumer confidence. us\n manufacturers''optimism is now almost as low as it was in 1991 - the last time\n the us economy experienced a recession. consumer confidence also fell sharply\n in january, driven by marked pessimism over the short - term future. quite why\n this break in confidence should have been so rapid is not easy to understand.\n and its origins will largely determine the nature of the us downturn. on the one\n hand, greater use of information technology to economise on inventories may have\n led to shorter lags between changes in final demand and changes in output. if\n so, then it is possible that the speed of the downturn will be matched by the\n speed of the recovery, leading to a short - lived episode of output growth close\n to zero as the result of an inventory correction. on the other hand, the slowdown\n could be much more protracted if the imbalances in the us economy which have built\n up in recent years start to unwind, leading to a reduction in spending as both\n households and businesses seek to reduce the amount of outstanding debt on their\n balance sheets. the key to the nature of the us downturn is what will happen to\n productivity growth. over the past five years there has been accumulating evidence\n that the application of information technology has raised productivity growth\n in the us economy - the \" new economy \". expectations of higher productivity growth\n increased demand by more than it raised supply initially, as firms invested in\n new technology and households anticipated higher future incomes. as a result,\n spending grew rapidly, outstripping supply and large imbalances emerged. the current\n account deficit in the us is now close to 5 % of gdp, a post - war record. it\n is sustainable as long as foreigners are prepared to finance it. so far, the profitability\n of'\n - 83 6. 06 27. 05 39. 42 37. 59 kccs ( no. in lakh ) 243. 07 271. 12 302. 35 337.\n 87 82. 43 gcc ( no. in lakh ) 13. 87 16. 99 21. 08 36. 29 22. 28 bc - ict accounts\n ( no. in lakh ) 132. 65 316. 30 573. 01 810. 38 677. 73 ict accounts - bc - total\n transactions ( no. in lakh ) 265. 15 841. 64 1410. 93 2546. 51 4799. 08 bis central\n bankers ’ speeches\n - core business and therefore set out to develop the expertise required to finance\n the smes successfully. fairly often the discussion on sme financing is reduced\n to two diametrically opposed positions. on one hand, smes are considered by banks\n as representing a high risk and therefore, should be avoided or only dealt with\n cautiously and at a premium price. on the other hand, banks are accused of being\n inflexible and risk averse and consequently irrelevant to the sector. it is important\n to understand where the truth lies in these two statements in order to advance\n the cause of sme financing. firstly, it is a fact that smes present higher credit\n risk than well - structured corporate entities. smes may not have proper accounting\n records, may have severe governance issues which undermine accountability, have\n poor access to markets, poor skill levels including financial illiteracy by promoters,\n lack collateral which the lender can rely on in the event of failure, may not\n even exist in an appropriate legal form and even the assessment of the viability\n of a project might be difficult. lending to smes can be a lenders nightmare for\n bankers. but it is also true that banks which are structured to deal with corporates\n are risk averse and inflexible when they deal with smes. often when they bring\n inappropriate risk assessment tools, they may focus too much on collateral rather\n than project viability. they may even regard sme financing as peripheral to their\n business. because of their limited knowledge of smes, they experience failure\n which itself reinforces the notion that smes are risky. what we want are financing\n institutions that are structured to respond to the unique characteristics of smes.\n specialised sme lending institutions are more likely to handle the risk problem\n presented by smes as a challenge to be overcome with appropriate products and\n credit risk management strategies and not as a basis for inaction or avoiding\n the sector altogether. in short, lending strategies which ensure success with\n corporates do not necessarily ensure similar success with smes. appropriate sme\n financing institutions must at the very least make lending to smes the core business.\n the second area of reflection that i would urge this meeting to consider is that\n of building appropriate financing models that have been shown to work in africa\n or other developing countries so that we all benefit from the best practices available.\n in south africa, for instance, franchising which allows the use of brand names\n and building capacity have been an bis central bankers ’ speeches important driver\n of sme financing since it reduces the perceived\n- source_sentence: and non - financial sectors take hold. these concerns are heightened\n by the recent upward shift in bond yields owing to the global \" higher - for -\n longer \" narrative and the flare - up of tensions in the middle east, which have\n added to the uncertainty surrounding the outlook. after a period of lower market\n volatility until august, the rising prospect of higher - forlonger rates has started\n to weigh on riskier asset valuations in recent months. risk sentiment in markets\n remains highly sensitive to further surprises in inflation and economic growth.\n higher than expected inflation or lower growth could trigger a rise in market\n volatility and risk premia, increasing the likelihood of credit events materialising.\n this brings me to the vulnerabilities in the non - bank financial sector. as regards\n credit risk, some non - banks remain heavily exposed to interest rate - sensitive\n sectors, such as highly indebted corporates and real estate. deteriorating corporate\n fundamentals and the ongoing correction in real estate markets could expose non\n - banks that have 2 / 4 bis - central bankers'speeches invested in these sectors\n to revaluation losses and investor outflows. furthermore, low levels of liquidity\n could expose investment funds to the potential risk of forced asset sales if macro\n - financial outcomes deteriorate. corporate profitability in the euro area has\n held up well, but higher interest rates are weighing on the debt servicing capacity\n of more vulnerable firms. a weakening economy could prove challenging for firms\n with high debt levels, subdued earnings and low interest coverage ratios. real\n estate firms are particularly vulnerable to losses stemming from the ongoing downturn\n in euro area commercial real estate markets. in an environment of tighter financing\n conditions and elevated uncertainty, real estate prices have declined markedly.\n the effects of higher interest rates have been compounded by structurally lower\n demand for some real estate assets following the pandemic. although banks'exposure\n to these markets is comparatively low, losses in this segment could act as an\n amplifying factor in the event of a wider shock. euro area households, especially\n those with lower incomes and in countries with mainly floatingrate mortgages,\n are being increasingly squeezed by the higher interest rates. tighter financing\n conditions have reduced the demand for housing, putting downward pressure on prices.\n on a more positive note, robust labour markets have so far supported household\n balance sheets, thereby mitigating the credit risk to banks. spreads in government\n bond markets have remained contained as many governments managed to secure cheap\n financing at longer maturities during the period of low interest rates.\n sentences:\n - market. and if we are to raise our national economic vantage point and take advantage\n of the momentum coming into 2013, thrift banks need to consider lending more to\n this economic segment. i say this because msmes provide great employment opportunities.\n a healthier msme sector will help ensure our economic growth is broad - based\n and inclusive. thrift banks, however, must not just lend more in terms of nominal\n amounts. the challenge to the industry really is to ensure that such lending continuously\n creates further opportunities. in this way, msmes can be assured of viability,\n regardless of mandated credit programs. final thoughts friends, you want this\n convention to be a discussion on broadening your horizon, our horizon. in trying\n to look ahead, i took us through the more scenic route of our recent past. the\n numbers tell us an encouraging story of growth and consistency of economic and\n financial stability. however, the paradox of financial stability is that we may\n just be at our weakest when we believe we are at our strongest position. complacency\n often sets in when positive news is continuous and this is reinforced by encouraging\n market parameters. if we fall into this trap of complacency, we risk losing our\n focus. for the thrift banking industry, the growth trajectory that many are anticipating\n suggests even better times ahead. it is therefore in our collective interest,\n if we wish to truly broaden our horizon, to agree on a common vision where thrift\n banks have a definitive role to play. bis central bankers ’ speeches in my remarks,\n i have suggested specific action points you may consider in order to accomplish\n just that, and over time broaden your horizon. i challenge the industry to 1 )\n further improve credit underwriting standards for real estate and consumer loans\n and raise your vantage point by looking at your processes with the lens of financial\n stability, 2 ) interact with your clients and raise your vantage point by enhancing\n your practices with the heart of consumer protection and financial education,\n and 3 ) increase lending to msmes and raise your vantage point by lending with\n the mind to create greater value. but much more can be done. as the old board\n takes its place in ctb lore, your new board is now at the helm to guide the industry\n forward. it is never an easy task to move forward but rest assured that the bangko\n sentral ng pilipinas will be with you as we take that journey together into broader\n horizons. thank you very much and good day to all of you. bis central bankers\n ’ speeches\n - prices, which may also come under upward pressure owing to adverse weather events\n and the unfolding climate crisis more broadly. most measures of underlying inflation\n continue to decline. the eurostat's flash estimate for inflation excluding energy\n and food points to a further decline to 4. 2 % in october, supported by improving\n supply conditions, the pass - through of previous declines in energy prices, as\n well as the impact of tighter monetary policy on demand and corporate pricing\n power. at the same time, domestic price pressures are still strong and 1 / 4 bis\n - central bankers'speeches are being increasingly driven by wage pressures and\n the evolution of profit margins. while most measures of longer - term inflation\n expectations stand around 2 %, some indicators remain elevated and need to be\n monitored closely. the resilience of the labour market has been a bright spot\n for the euro area economy, but there are signs that the labour market is beginning\n to weaken. fewer new jobs are being created and, according to the latest flash\n estimate, employment expectations have continued to decline in october for both\n services and manufacturing. monetary policy based on our assessment of the inflation\n outlook, the dynamics of underlying inflation and the strength of monetary policy\n transmission, the governing council decided to keep the three key ecb interest\n rates unchanged at its october meeting. the incoming information has broadly confirmed\n our previous assessment of the medium - term inflation outlook. our past interest\n rate increases continue to be transmitted forcefully into financial and monetary\n conditions. banks'funding costs have continued to rise and are being passed on\n to businesses and households. the combination of higher borrowing rates and weakening\n activity led to a further sharp drop in credit demand in the third quarter of\n this year. and credit standards have tightened again. we are also seeing increasing\n signs of the impact of our policy decisions on the real economy. further tightening\n is still in the pipeline from the current policy stance, and it is set to further\n dampen demand and help push down inflation. we are determined to ensure that inflation\n returns to our 2 % medium - term target in a timely manner. based on our current\n assessment, we consider that the key ecb interest rates are at levels that, maintained\n for a sufficiently long duration, will make a substantial contribution to this\n goal. we will continue to follow a data - dependent approach to determining the\n appropriate level and duration of restriction. financial stability let me now\n turn to financial stability. in our upcoming financial stability review, we highlight\n that the financial stability outlook remains fragile as the gradual effects of\n tighter financial conditions on both the financial\n - the conservation of natural resources. it was only in the fitness of things that\n he delivered the inaugural address today. sir, we are overwhelmed by your insightful\n address and i am sure all of us have immensely benefited listening to his valuable\n views on a subject which is not only very dear to you but also so vital for the\n country ’ s economic growth. 4. i also note that the seminar includes several\n eminent speakers with vast experience on the subject. there are sessions on agricultural\n productivity, role of research and technology in improving agricultural productivity,\n linkages between productivity and farm income, incentivizing productivity enhancing\n investments, relation between credit growth and productivity growth and mitigation\n of risks in agriculture. in my address today, coming as do from the rbi, i would\n broadly focus on major trends / issues in agricultural productivity and credit,\n and the role of agricultural credit in improving agricultural productivity. i\n would also touch upon some of the steps that can yield results in the short -\n term. importance of agriculture 5. as you are aware, the recent indian growth\n story has been service - led. services sector has completely replaced agriculture,\n which was traditionally the largest contributor to india ’ s gdp. however, the\n fact that agriculture has the smallest share in gdp of only about 14 per cent\n today from a high of more than 50 per cent, does not belittle its importance for\n bis central bankers ’ speeches the indian economy. this is because first, as we\n all know, agriculture remains the largest employer having a share of around 60\n per cent. secondly, it holds the key to creation of demand in other sectors and\n remains by far an important indirect contributor to india ’ s gdp growth. the\n agriculture sector needs to grow at least by 4 per cent for the economy to grow\n at 9 per cent. thus, though having a small share, the fluctuations in agricultural\n production can have large and significant impact on overall gdp growth. thirdly,\n since food is an important component in basket of commodities used for measuring\n consumer price indices, it is necessary that food prices are maintained at reasonable\n levels to ensure food security, especially for the deprived sections of our society.\n in fact, food security is emerging as an important policy concern, and the role\n of agriculture in ensuring equitable access to food has added a new perspective\n for policy makers. trends in agricultural productivity 6. i would like to discuss\n certain trends in agricultural productivity in india. as is wellknown, the year\n 1968 marked the beginning of a turning point in indian agriculture. the country\n- source_sentence: having a reserve currency may also be desirable for other reasons.\n first, with more reserve currencies available, portfolio diversification opportunities\n are enhanced. this is desirable because, all else equal, it would allow investors\n to move further out on the risk / return frontier. second, more currencies are\n likely to be close substitutes, which could dampen currency volatility. with many\n viable reserve currencies available, no particular one would necessarily have\n to bear the bulk of any adjustment. the dollar ’ s dominant reserve currency status\n has sometimes been referred to as the united states ’ “ exorbitant privilege,\n ” implying that the u. s. benefits extraordinarily from this privileged status.\n i ’ d argue that the situation is much more nuanced. yes, this status does allow\n the u. s. to benefit from seigniorage. more than half of all u. s. currency outstanding\n is held abroad. but, there are also costs of being the dominant reserve currency.\n for example, this can lead to shifts in the valuation of the dollar that are due\n primarily to developments abroad that affect risk appetites and international\n capital flows. in such cases, the dollar ’ s valuation can be pushed to levels\n inconsistent with u. s. economic fundamentals. for the united states, i believe\n that the most important goal must be to keep our own house in order. if we do\n this, then i expect that the u. s. dollar will earn the right to remain the most\n important reserve currency in the world. the united states has a number of advantages\n in sustaining the dominant reserve currency status of the u. s. dollar. first,\n there is a first - mover advantage. as the leading reserve currency in the world,\n there is no strong incentive for countries to move to other currencies as long\n as the dollar continues to have the attributes i discussed earlier. the history\n of reserve currency usage is characterized by considerable inertia. the u. s.\n dollar emerged as the leading reserve currency quite a bit after it became the\n world ’ s largest economy. typically, the loss of dominant reserve currency status\n requires either substantial economic decline or political instability that motivates\n foreign counterparties to shift to a new reserve currency. second, the u. s. has\n the deepest and most liquid capital markets in the world. this is important in\n making u. s. treasuries and agency mortgage - backed securities attractive holdings\n as part of countries ’ foreign exchange reserve portfolio\n sentences:\n - as those on student visas, are assumed to begin arriving early next year, subject\n to appropriate quarantine restrictions. but tourists and other short - term visitors\n will not be able to return until later. in the baseline and upside scenarios,\n we assume that the borders reopen in mid 2021. in the downside scenario, where\n the global spread of the virus does not subside as quickly, we assume the borders\n are closed for all of 2021. it is always possible to construct other scenarios.\n the near term would be stronger if there were a major medical breakthrough on\n treatments soon. an effective vaccine would take a bit longer to be distributed,\n so it would mainly affect outcomes next year and the year after. but it could\n also result in a stronger recovery than we have assumed even in the upside scenario\n presented here. a worse outcome than our downside could be conceivable if the\n virus cannot be contained and further waves of infection occur around the world\n for some years yet. there are also plenty of other risks that can be contemplated\n ; we discuss some of these in the statement. geopolitical tensions were an issue\n even before the coronavirus outbreak, and could escalate further. the pandemic\n has also in some places exacerbated domestic political tensions. it is hard to\n know how these tensions will play out over the next couple of years. if they escalate,\n it is possible that some countries'recoveries will be derailed. domestically,\n there are also a number of uncertainties that go beyond the direct effects of\n the virus and associated activity restrictions. for example, we have assumed that\n households and businesses are quite cautious in their use of the fiscal and other\n cash flow support they have been receiving. it is possible that people spend more\n out of that support than we are assuming. it is even possible that some people\n do more to make up for the consumption opportunities that were not available during\n periods of lockdown and other activity restrictions. on the downside, the longer\n the economy remains weak, the more the recovery will be impeded by scarring effects\n on workers, the destruction of business supply networks and other lingering damage.\n what has changed in the past three months the scenarios presented in the past\n few months. statement incorporate several lessons from the experience of the first,\n the economic contractions induced by health - related restrictions on activity\n in the june quarter were very large. however, they were not quite as severe as\n initially expected. the peak - to - trough declines in output and hours worked\n - '. norway was hit by oil price shocks in the following years and economic reforms\n were implemented in several areas. after a period it became clear that the central\n government budget would be in surplus and that transfers would be made to the\n fund. norway ’ s experience from its first 30 years as an oil - producing nation\n led to the introduction of the fiscal rule, which has been a key element of norwegian\n economic policy for the past decade. report no. 29 to the storting of 2001 laid\n down guidelines for the phasing - in of petroleum revenues into the norwegian\n economy, establishing two main principles : economic policy must contribute to\n stable economic developments and be sustainable over time. by linking petroleum\n revenue spending to the expected real return on the fund – and not to current\n petroleum revenues – the fiscal rule provided for a gradual and sustainable phasingin\n of the revenues. if we can restrict spending to the return, the fund will never\n shrink. norway will also be less vulnerable to fluctuations in current petroleum\n revenues. ( chart 8 : effect on the size of gpfg of change in oil price and return\n ) in the first ten years after the establishment of the fiscal rule, the pace\n of the fund ’ s growth was determined by oil prices. prospects in 2001 indicated\n that a 25 percent higher oil price over a 10 - year period would increase the\n size of the fund by almost nok 800 billion. uncertainty with regard to the return\n on the fund was far less important for growth. this situation will change as the\n fund grows and oil and gas production declines. from 2020 to 2030, the oil price\n will play a less prominent role for the size of the fund, while the return on\n the fund will be all the more important. in the initial years, the actual return,\n adjusted for inflation and costs, was close to 4 percent. in recent years, with\n the financial crisis and the sovereign debt crisis, the real return has been lower,\n averaging about 2½ percent since 1998. source : lie, e. and c. venneslan : over\n evne. finansdepartementet 1965 – 1992 [ beyond our power. ministry of finance\n 1965 – 1992 ]. pax forlag, 2010. see report no. 25 to the storting ( 1973 - 74\n ) : “ petroleumsvirksomhetens plass i det norske samfunn ” [ the role of petroleum\n activity in norwegian society ]. bis central bankers ’ speeches we should be careful\n about taking a rear -'\n - ', increasingly relies on price signals generated by trading activity that takes\n place daily in these markets. the reliance on secondary market trading for price\n discovery constitutes the fundamental difference between funds from securities\n markets and loans from banks. let me be a bit more specific. in securities markets,\n investment decisions are driven by prices that arise from a trading process that\n reconciles differential information from a diverse group of investors. in bank\n loans, investment decisions are based on the bank ’ s private information about\n specific borrowers. while a bank makes its own investment decisions, securities\n markets rely on the consensus of a multitude of investors. when securities markets\n work well, they provide efficient ways of aggregating information and allocating\n risks among a wide range of investors. in order to function well, however, these\n markets require a trading infrastructure. this infrastructure may consist of an\n exchange, a network of brokers and dealers, and a clearing system. these markets\n also rely on a cadre of relatively well - informed investors, who confidently\n judge asset prices and take positions on the – 2 – strength of their judgments.\n if the trading infrastructure fails or investors lose confidence, trading will\n grind to a halt. the global fixed - income markets are unlike equity markets.\n in equity markets, everyone knows something about the trading infrastructure,\n which is centralized in exchanges. thus, there is no question as to the focal\n point of trading information. but the importance of fixed - income markets, which\n are multiple - dealer, over - the - counter markets, is sometimes hard to appreciate\n because they are so decentralized. in the united states, the bond market is where\n companies have been raising most of their funds in recent years. during the last\n ten years, for example, u. s. nonfinancial corporations borrowed a net amount\n of $ 785 billion in the form of bonds, three times the net amount they borrowed\n from banks. over this same period, these companies as a group spent $ 600 billion\n more to retire stock – through buybacks and mergers – than they raised in new\n offerings. accompanying these increased levels of debt market activity has been\n a continuous process of financial innovation. this innovation has served to unbundle\n different kinds of risk and, thereby, to enlarge the menu of risks that investors\n may choose to bear. for example, interest - rate swaps, futures and options help\n reconfigure various interest - rate risks. total return swaps and creditspread\n options are tools for reallocating the payment risks primarily of emerging market'\n- source_sentence: education shocks arising from the pandemic could signify the largest\n reversal in human development on record, equivalent to erasing all the progress\n in human development of the past six years4. the risk of reversing progress in\n achieving financial inclusion is accentuating why financial inclusion matters\n now more than ever. optimising islamic finance for inclusive economic recovery\n policy and regulatory responses across the globe, and likewise in malaysia, have\n been focused on safeguarding economic resilience, managing risks to financial\n stability and minimising repercussions to society. this is done in tandem with\n the different stages of the pandemic – namely, containment, stabilisation and\n recovery. at the onset of the crisis, governments along with financial regulators\n have deployed sizeable stimulus packages and various assistance programmes in\n order to contain the crisis and stabilise the economy. this includes addressing\n demand and supply disruptions, maintaining cash flows and keeping workers employed.\n in malaysia, the total stimulus package amounted to usd 73. 55 billion ( rm3056\n billion ) with an additional fiscal injection by the government totalling 1 /\n 4 bis central bankers'speeches rm45 billion. as at september 2020, a total of\n 2. 63 million workers and 321, 000 employers had benefitted from the wage subsidy\n programme, involving an expenditure of rm10. 4 billion. to provide further stimulus\n to the economy, the bank has reduced the overnight policy rate ( opr ) by a cumulative\n 125 basis points ( from 3. 00 % to 1. 75 % ) this year, alongside reduction in\n the statutory reserve requirement by 100 basis points ( from 3. 00 % to 2. 00\n % ). the reduction in the opr is intended to provide additional policy stimulus\n to accelerate the pace of economic recovery. the financial industry, including\n islamic financial institutions, also lent support to their borrowers and customers.\n in the first half of 2020, a total of rm120 billion7 was disbursed in lending\n / financing to smes, with more accounts being approved8 in aggregate in 2020 compared\n to the same period in previous years. islamic financial institutions and related\n associations have been actively educating and reaching out to affected borrowers\n about the financial assistance programmes available in response to the pandemic.\n the takaful and insurance industry also facilitated affected certificate holders\n by offering temporary deferment of contribution and premium to promote continuity\n of takaful protection coverage. more than 1. 1 million9 certificate and policyholders\n have benefited from this relief measure. while the\n sentences:\n - education at an early age. a credit union is similar to a commercial bank, it\n can provide the same products and services like a commercial bank, but what is\n different is that the credit union is owned by members who are the shareholders\n and at the same time are customers to the credit union. credit unions can provide\n different saving accounts to suit the needs of their members. they can create\n saving products such as junior saving accounts for the children. credit unions\n provide lending products for their members. and credit unions can provide insurance\n coverage for their members as well. one of the reasons people explain why they\n don ’ t want to open an account with a commercial bank is that because banks charge\n fees to keep their accounts. while paying fees is something that we all cannot\n avoid because provision of banking services costs money, in a credit union the\n interests, fees and charges that are paid by members is distributed back to the\n members as net interest income. i say this because i am also member of our staff\n credit union called bokolo credit union limited day one when i join the central\n bank of solomon islands and know the benefit credit unions can give members. as\n the registrar for credit unions, i take note of the progress in the growth of\n the assets of both the soltuna employees credit union limited and the tuna trust\n credit union limited. if the board and management of these credit unions continue\n to manage these credit unions professionally, they will both be the vehicle to\n encourage our children and youth here in noro and munda community to learn how\n to save and earn and become good financial and economic citizens of our country.\n finally, let me take this opportunity to thank the many people who made this global\n money week celebration possible. to the board of directors of both tuna trust\n credit union limited and soltuna employees credit union limited. i salute you\n for accepting our request to work together to host this global money week. i would\n like to thank selwyn talasasa, one of the long serving credit union promoters\n who has been a credit union person since day one of his career. he has been very\n helpful in organizing our global money week. thank you selwyn. i also thank the\n global money week committee for their assistance in organising this global money\n week. and thanks to the management of both national fisheries development limited\n and soltuna limited for your support in hosting this global money week. thank\n you nfd and soltuna, you made us all proud as good cooperate citizens. to our\n noro town clerk for your support.\n - 'in tandem, the policy rate of this central bank would have to be changed frequently\n and forcefully in the same direction to offset the shock. but what would happen\n in the other economy, if it – unlike the first – were more prone to supply shocks?\n experience suggests that supply shocks yield sharp transitory increases in inflation,\n possibly followed by smaller, more permanent “ secondround ” effects, though the\n longer - run impact on inflation is obviously significantly determined by the\n response of monetary policy. given the transitory nature of the initial inflation\n bursts, the simple hypothetical rule – which incorporates the reaction to expected\n inflation – would advise the central bank to “ look through ” the immediate disturbance\n and change policy only to the extent needed to offset the anticipated more permanent\n effects of the shock on inflation in subsequent quarters. its policy rate, again,\n would be observed to be less variable. what is important to note is that the same\n rule – equally active strategies – would support two different patterns of observed\n policy behaviour in different economic environments. the third case is perhaps\n the most interesting of all. here exogenous shocks are identical, but economic\n structures differ. different transmission mechanisms therefore propagate the same\n shocks with lags that vary between the two economies. the first economy has more\n rigid adjustment mechanisms : price - setters and wage - negotiators are more\n sluggish than those in the other economy in processing economic news – including\n changes in the stance of monetary policy – and bringing them to bear on their\n decisions. what is the source of those rigidities in the first economy? there\n can be many reasons for rigidity. perhaps labour practices and contractual institutions\n – dating from the early post - war decades when the economy was heavily regulated\n – induce distortions in large a similar interpretation of “ interest rate inertia\n ” – the tendency of central banks to adjust rates in the same direction and in\n small steps – can be found in g. rudebusch, “ term structure evidence on interest\n rate smoothing and monetary policy inertia ”, journal of monetary economics, vol.\n 49 ( pp. 1161 - 1187 ), 2002. segments of the labour market. this stands in the\n way of an efficient matching of skills and productive capabilities. perhaps tight\n regulatory restraint on business and statutory inhibitions discourage innovation\n and impede a faster response to new shocks and new opportunities. whatever the\n source of rigidity, the observational result is that prices and wages in the first\n economy reflect changes in fundamentals with considerable lags. how should'\n - 'fund ( uncdf ) are also actively developing financial inclusion knowledge, and\n supporting countries in implementing financial inclusion strategies globally.\n over the years, we have since witnessed the rapid deployment of innovative financial\n inclusion solutions that have changed the lives of many. since the introduction\n of the mpesa mobile banking service in kenya, the number of adults with access\n to financial services has increased from 42 % in 2011 to 75 % today. agent banking\n in malaysia has provided consumers with an innovative and alternative channel\n to access financial services, resulting in the percentage of sub - districts served\n by financial services access point to increase from 46 % in 2011 to 96 % in 2014.\n this has enabled 99 % of malaysians, particularly those in rural areas, to conveniently\n access and benefit from financial services. the microfinance “ global financial\n development report 2014 ”, world bank, 2014. discussion note “ redistribution,\n inequality, and growth ”, imf, 2014. bis central bankers ’ speeches institution\n grameen bank in bangladesh has extended credit through its 2, 500 branches to\n almost 8 million people, with 60 % of them lifted from poverty. the loan recovery\n rate is higher than 98 percent. the experience in bangladesh, in particular, dispels\n the myth that the poor is not bankable. of proportionate regulation and the global\n standards the proportionate application of global standards for financial regulation\n is a critical factor in enabling innovative financial inclusion solutions, ensuring\n its delivery in a safe and sound manner. my remarks today will touch on the following\n topics : recent developments in global standards and financial inclusion ; focus\n areas to advance proportionality in practice ; and the importance of the global\n symposium to galvanise action. recent developments in global standards and financial\n inclusion global standards developed after the global financial crisis were designed\n to address the financial stability and integrity issues that came to the fore\n in developed countries. many of these issues involved the activities of global\n systemically important financial institutions. the effects of these standards\n on financial inclusion, such as the impact on smaller financial institutions in\n developing countries most likely were not taken into consideration. fortunately,\n standard - setting bodies have now recognised the principle of proportionality,\n emphasising on the balance between objectives of financial stability, integrity\n and inclusion. however, the real challenge lies in implementing proportionality\n in practice. if principles are not implemented in a proportionate manner, there\n could be unintended consequences for financial inclusion. for example, despite\n the financial action task force ’ s guidance on the implementation of a risk -\n based approach to anti money laundering /'\nmodel-index:\n- name: SentenceTransformer based on samchain/EconoBert\n results:\n - task:\n type: semantic-similarity\n name: Semantic Similarity\n dataset:\n name: econosentencev2\n type: econosentencev2\n metrics:\n - type: pearson_cosine\n value: 0.8556254615515724\n name: Pearson Cosine\n - type: spearman_cosine\n value: 0.8619885397301873\n name: Spearman Cosine\n - task:\n type: semantic-similarity\n name: Semantic Similarity\n dataset:\n name: econosentence prior df\n type: econosentence-prior_df\n metrics:\n - type: pearson_cosine\n value: 0.8373175950726144\n name: Pearson Cosine\n - type: spearman_cosine\n value: 0.8294930387036759\n name: Spearman Cosine\n---\n\n# SentenceTransformer based on samchain/EconoBert\n\nThis is a [sentence-transformers](https://www.SBERT.net) model finetuned from [samchain/EconoBert](https://huggingface.co/samchain/EconoBert). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.\nThis model is the second version of the previous [samchain/econo-sentence-v1](https://huggingface.co/samchain/econo-sentence-v1). They both used the same base model but have been trained on different datasets.\n\nThe model has been evaluated both on the newest dataset and its precedent version:\n- `econo-pairs-v2` is the newest dataset. Check its card [here](https://huggingface.co/samchain/econo-pairs-v2)\n- `econo-pairs-v1`is the oldest version. Check its card [here](https://huggingface.co/samchain/econo-pairs)\n \n\n## Model Details\n\n### Model Description\n- **Model Type:** Sentence Transformer\n- **Base model:** [samchain/EconoBert](https://huggingface.co/samchain/EconoBert) <!-- at revision 1554ddcdc25e1886cc43e05b9c77a2b8c4888da6 -->\n- **Maximum Sequence Length:** 512 tokens\n- **Output Dimensionality:** 768 dimensions\n- **Similarity Function:** Cosine Similarity\n<!-- - **Training Dataset:** Unknown -->\n<!-- - **Language:** Unknown -->\n<!-- - **License:** Unknown -->\n\n### Model Sources\n\n- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)\n- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)\n- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)\n\n### Full Model Architecture\n\n```\nSentenceTransformer(\n (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel \n (1): Pooling({'word_embedding_dimension': 768, '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})\n)\n```\n\n## Usage\n\n### Direct Usage (Sentence Transformers)\n\nFirst install the Sentence Transformers library:\n\n```bash\npip install -U sentence-transformers\n```\n\nThen you can load this model and run inference.\n```python\nfrom sentence_transformers import SentenceTransformer\n\n# Download from the 🤗 Hub\nmodel = SentenceTransformer(\"samchain/econosentence-v2\")\n# Run inference\nsentences = [\n \"education shocks arising from the pandemic could signify the largest reversal in human development on record, equivalent to erasing all the progress in human development of the past six years4. the risk of reversing progress in achieving financial inclusion is accentuating why financial inclusion matters now more than ever. optimising islamic finance for inclusive economic recovery policy and regulatory responses across the globe, and likewise in malaysia, have been focused on safeguarding economic resilience, managing risks to financial stability and minimising repercussions to society. this is done in tandem with the different stages of the pandemic – namely, containment, stabilisation and recovery. at the onset of the crisis, governments along with financial regulators have deployed sizeable stimulus packages and various assistance programmes in order to contain the crisis and stabilise the economy. this includes addressing demand and supply disruptions, maintaining cash flows and keeping workers employed. in malaysia, the total stimulus package amounted to usd 73. 55 billion ( rm3056 billion ) with an additional fiscal injection by the government totalling 1 / 4 bis central bankers'speeches rm45 billion. as at september 2020, a total of 2. 63 million workers and 321, 000 employers had benefitted from the wage subsidy programme, involving an expenditure of rm10. 4 billion. to provide further stimulus to the economy, the bank has reduced the overnight policy rate ( opr ) by a cumulative 125 basis points ( from 3. 00 % to 1. 75 % ) this year, alongside reduction in the statutory reserve requirement by 100 basis points ( from 3. 00 % to 2. 00 % ). the reduction in the opr is intended to provide additional policy stimulus to accelerate the pace of economic recovery. the financial industry, including islamic financial institutions, also lent support to their borrowers and customers. in the first half of 2020, a total of rm120 billion7 was disbursed in lending / financing to smes, with more accounts being approved8 in aggregate in 2020 compared to the same period in previous years. islamic financial institutions and related associations have been actively educating and reaching out to affected borrowers about the financial assistance programmes available in response to the pandemic. the takaful and insurance industry also facilitated affected certificate holders by offering temporary deferment of contribution and premium to promote continuity of takaful protection coverage. more than 1. 1 million9 certificate and policyholders have benefited from this relief measure. while the\",\n 'education at an early age. a credit union is similar to a commercial bank, it can provide the same products and services like a commercial bank, but what is different is that the credit union is owned by members who are the shareholders and at the same time are customers to the credit union. credit unions can provide different saving accounts to suit the needs of their members. they can create saving products such as junior saving accounts for the children. credit unions provide lending products for their members. and credit unions can provide insurance coverage for their members as well. one of the reasons people explain why they don ’ t want to open an account with a commercial bank is that because banks charge fees to keep their accounts. while paying fees is something that we all cannot avoid because provision of banking services costs money, in a credit union the interests, fees and charges that are paid by members is distributed back to the members as net interest income. i say this because i am also member of our staff credit union called bokolo credit union limited day one when i join the central bank of solomon islands and know the benefit credit unions can give members. as the registrar for credit unions, i take note of the progress in the growth of the assets of both the soltuna employees credit union limited and the tuna trust credit union limited. if the board and management of these credit unions continue to manage these credit unions professionally, they will both be the vehicle to encourage our children and youth here in noro and munda community to learn how to save and earn and become good financial and economic citizens of our country. finally, let me take this opportunity to thank the many people who made this global money week celebration possible. to the board of directors of both tuna trust credit union limited and soltuna employees credit union limited. i salute you for accepting our request to work together to host this global money week. i would like to thank selwyn talasasa, one of the long serving credit union promoters who has been a credit union person since day one of his career. he has been very helpful in organizing our global money week. thank you selwyn. i also thank the global money week committee for their assistance in organising this global money week. and thanks to the management of both national fisheries development limited and soltuna limited for your support in hosting this global money week. thank you nfd and soltuna, you made us all proud as good cooperate citizens. to our noro town clerk for your support.',\n 'fund ( uncdf ) are also actively developing financial inclusion knowledge, and supporting countries in implementing financial inclusion strategies globally. over the years, we have since witnessed the rapid deployment of innovative financial inclusion solutions that have changed the lives of many. since the introduction of the mpesa mobile banking service in kenya, the number of adults with access to financial services has increased from 42 % in 2011 to 75 % today. agent banking in malaysia has provided consumers with an innovative and alternative channel to access financial services, resulting in the percentage of sub - districts served by financial services access point to increase from 46 % in 2011 to 96 % in 2014. this has enabled 99 % of malaysians, particularly those in rural areas, to conveniently access and benefit from financial services. the microfinance “ global financial development report 2014 ”, world bank, 2014. discussion note “ redistribution, inequality, and growth ”, imf, 2014. bis central bankers ’ speeches institution grameen bank in bangladesh has extended credit through its 2, 500 branches to almost 8 million people, with 60 % of them lifted from poverty. the loan recovery rate is higher than 98 percent. the experience in bangladesh, in particular, dispels the myth that the poor is not bankable. of proportionate regulation and the global standards the proportionate application of global standards for financial regulation is a critical factor in enabling innovative financial inclusion solutions, ensuring its delivery in a safe and sound manner. my remarks today will touch on the following topics : recent developments in global standards and financial inclusion ; focus areas to advance proportionality in practice ; and the importance of the global symposium to galvanise action. recent developments in global standards and financial inclusion global standards developed after the global financial crisis were designed to address the financial stability and integrity issues that came to the fore in developed countries. many of these issues involved the activities of global systemically important financial institutions. the effects of these standards on financial inclusion, such as the impact on smaller financial institutions in developing countries most likely were not taken into consideration. fortunately, standard - setting bodies have now recognised the principle of proportionality, emphasising on the balance between objectives of financial stability, integrity and inclusion. however, the real challenge lies in implementing proportionality in practice. if principles are not implemented in a proportionate manner, there could be unintended consequences for financial inclusion. for example, despite the financial action task force ’ s guidance on the implementation of a risk - based approach to anti money laundering /',\n]\nembeddings = model.encode(sentences)\nprint(embeddings.shape)\n# [3, 768]\n\n# Get the similarity scores for the embeddings\nsimilarities = model.similarity(embeddings, embeddings)\nprint(similarities.shape)\n# [3, 3]\n```\n\n<!--\n### Direct Usage (Transformers)\n\n<details><summary>Click to see the direct usage in Transformers</summary>\n\n</details>\n-->\n\n<!--\n### Downstream Usage (Sentence Transformers)\n\nYou can finetune this model on your own dataset.\n\n<details><summary>Click to expand</summary>\n\n</details>\n-->\n\n<!--\n### Out-of-Scope Use\n\n*List how the model may foreseeably be misused and address what users ought not to do with the model.*\n-->\n\n## Evaluation\n\n### Metrics\n\n#### Semantic Similarity\n\n* Datasets: `econo-pairs-v2` and `econo-pairs-v1`\n* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)\n\nEcono-sentence-v2 evaluation results:\n\n| Metric | econo-pairs-v2 | econo-pairs-v1 |\n|:--------------------|:----------------|:-----------------------|\n| pearson_cosine | 0.8556 | 0.8373 |\n| **spearman_cosine** | **0.862** | **0.8295** |\n\nEcono-sentence-v1 evaluation results:\n\n| Metric | econo-pairs-v2 | econo-pairs-v1 |\n|:--------------------|:----------------|:-----------------------|\n| pearson_cosine | 0.7642 | 0.8999 |\n| **spearman_cosine** | **0.758** | **0.8358** |\n\nThe v2 and v1 differs in terms of training as the v2 has been train on 0.5 labels, while the v1 is only trained on 1 and 0.\nThe results show that the v2 still performs on par with its predecessor on the prior dataset while improving the performance by 10 points on the latest. \n\n<!--\n## Bias, Risks and Limitations\n\n*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*\n-->\n\n<!--\n### Recommendations\n\n*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*\n-->\n\n## Training Details\n\n### Training Dataset\n\n#### Train set of econo-pairs-v2\n\n* Size: 45,044 training samples\n* Columns: <code>text1</code>, <code>text2</code>, and <code>label</code>\n* Approximate statistics based on the first 1000 samples:\n | | text1 | text2 | label |\n |:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:---------------------------------------------------------------|\n | type | string | string | float |\n | details | <ul><li>min: 3 tokens</li><li>mean: 454.6 tokens</li><li>max: 505 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 458.88 tokens</li><li>max: 505 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.49</li><li>max: 1.0</li></ul> |\n* Samples:\n | text1 | text2 | label |\n |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------|\n | <code>in credit intermediation is not disrupted. i hope the analysis and observations i have shared today can assist with a process of learning, evolving, and improving that will prevent future disruptions. thank you. bis central bankers ’ speeches bis central bankers ’ speeches bis central bankers ’ speeches</code> | <code>richard byles : update and outlook for the jamaican economy monetary policy press statement by mr richard byles, governor of the bank of jamaica, at the quarterly monetary policy report press conference, kingston, 21 february 2024. * * * introduction good morning and welcome to our first quarterly monetary policy report press conference for 2024. the year has started with inflation being above the bank's inflation target. statin reported last week that headline inflation at january 2024 was 7. 4 per cent, which is higher than the outturns for the previous three months, and above the bank's target of 4 to 6 per cent. core inflation, however, which excludes food and fuel prices from the consumer price index, was 5. 9 per cent, which is lower than the 7. 1 per cent recorded in january 2023. the inflation outturn at january is higher than the bank had projected in november 2023. as we communicated then, in the wake of the announcement by the minister of finance and the public service of th...</code> | <code>0.0</code> |\n | <code>european central bank : press conference - introductory statement introductory statement by mr jean - claude trichet, president of the european central bank, frankfurt am main, 4 may 2006. * * * ladies and gentlemen, let me welcome you to our press conference and report on the outcome of today ’ s meeting of the ecb ’ s governing council. the meeting was also attended by commissioner almunia. on the basis of our regular economic and monetary analyses, we have decided to leave the key ecb interest rates unchanged. overall, the information which has become available since our last meeting broadly confirms our earlier assessment of the outlook for price developments and economic activity in the euro area, and that monetary and credit growth remains very dynamic. against this background, the governing council will exercise strong vigilance in order to ensure that risks to price stability over the medium term do not materialise. such vigilance is particularly warranted in a context of ample...</code> | <code>vitor constancio : strengthening european economic governance – surveillance of fiscal and macroeconomic imbalances speech by mr vitor constancio, vice - president of the european central bank, at the brussels economic forum, brussels, 18 may 2011. * * * thank you very much for the invitation to participate in this panel today. the session is intended to focus on “ surveillance of fiscal and macroeconomic imbalances ”, which arguably is the most important strand of the economic governance package. but one should consider that this package – beyond the 6 legislative acts – also contains three other strands : the euro plus pact, the creation of the european stability mechanism and the setting - up of the european system of financial supervision. a fundamental strengthening of economic governance in the euro area requires simultaneous progress in all three areas. the first strand is necessary to prevent and correct imbalances ; the second to ensure the conditions for future growth and com...</code> | <code>0.5</code> |\n | <code>. innovation is already altering the power source of motor vehicles, and much research is directed at reducing gasoline requirements. at present, gasoline consumption in the united states alone accounts for 11 percent of world oil production. moreover, new technologies to preserve existing conventional oil reserves and to stabilize oil prices will emerge in the years ahead. we will begin the transition to the next major sources of energy perhaps before midcentury as production from conventional oil reservoirs, according to central tendency scenarios of the energy information administration, is projected to peak. in fact, the development and application of new sources of energy, especially nonconventional oil, is already in train. nonetheless, it will take time. we, and the rest of the world, doubtless will have to live with the uncertainties of the oil markets for some time to come.</code> | <code>oil industry to build inventories, demand from investors who have accumulated large net long positions in distant oil futures and options is expanding once again. such speculative positions are claims against future oil holdings of oil firms. currently, strained capacity has limited the ability of oil producers to quickly satisfy this markedly increased demand for inventory. adding to the difficulties is the rising consumption of oil, especially in china and india, both of which are expanding economically in ways that are relatively energy intensive. even the recent notable pickup in opec output, by exhausting most of its remaining excess capacity, has only modestly satisfied overall demand. output from producers outside opec has also increased materially, but investment in new producing wells has lagged, limiting growth of production in the near term. crude oil prices are also being distorted by shortages of capacity to upgrade the higher sulphur content and heavier grades of crude oi...</code> | <code>1.0</code> |\n* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:\n ```json\n {\n \"scale\": 20.0,\n \"similarity_fct\": \"pairwise_cos_sim\"\n }\n ```\n\n### Evaluation Dataset\n\n#### Test set of econo-pairs-v2\n\n* Size: 11,261 evaluation samples\n* Columns: <code>text1</code>, <code>text2</code>, and <code>label</code>\n* Approximate statistics based on the first 1000 samples:\n | | text1 | text2 | label |\n |:--------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:--------------------------------------------------------------|\n | type | string | string | float |\n | details | <ul><li>min: 4 tokens</li><li>mean: 451.65 tokens</li><li>max: 505 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 457.79 tokens</li><li>max: 505 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.5</li><li>max: 1.0</li></ul> |\n* Samples:\n | text1 | text2 | label |\n |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------|\n | <code>the drying up of the unsecured interbank market a price worth paying for lower interest rate volatility and tighter interest rate control in the money markets? at present, it is difficult to envisage how the unsecured interbank money market could be revived in full in the near future, reaching the pre - global financial crisis levels. this is due both to regulatory reasons and to fragmentation. post - crisis, banks appear reluctant to lend each other in the unsecured interbank market, pricing such lending at levels that are sufficient to compensate for counterparty credit risk. at the same time, short - term ( below 6 months ) interbank funding does not contribute towards satisfying the nsfr requirements. as a result, the unsecured segment of the interbank market is unattractive for prospective borrowers and is not expected to regain importance. furthermore, heterogeneity within and across the banking systems of euro area countries impedes the efficient redistribution of reserves acros...</code> | <code>##obalisation, artificial intelligence, as well as the green transition and climate change adaptation policies. 1 / 7 bis - central bankers'speeches given the uncertain overall impact of these effects and the significant unknowns and ambiguity surrounding the computation of natural interest rates, their prospective level and evolution are not easily determined. turning now to the lessons learnt, experience gained during the past crises has been extremely valuable. first, monetary policy needs to remain flexible so that it can accommodate potential future shocks to price stability, of any nature and direction. second, financial stability considerations need to be taken into account in the decision making of monetary policymakers, in their pursuit of price stability. the past decade provided us with new insights into what monetary policy can do. we are now better prepared to implement the appropriate monetary policy mix, with standard and non - standard tools, if we were to face new chal...</code> | <code>1.0</code> |\n | <code>therefore worries me that greece might go backwards in terms of its reform agenda. suffice to say, it is not enough to correct misalignments solely at the national level. we also need structural reforms at the european level. and here, a lot has been done since the crisis broke out. first, the rules of the stability and growth pact were tightened and a fiscal compact was adopted in order to restore confidence in public finances. second, a procedure for identifying macroeconomic imbalances at an early stage was established. and third, a crisis mechanism was set up to serve as a “ firewall ”, safeguarding the stability of the financial system in the euro area. in addition to these measures, the euro area took a major step toward deeper financial integration. on 4 november 2014, the first pillar of a european banking union was put in place. on that day, the ecb assumed responsibility for supervising the 123 largest banks in the euro area. together, the banks concerned account for more tha...</code> | <code>andreas dombret : the euro area – where do we stand? speech by dr andreas dombret, member of the executive board of the deutsche bundesbank, at the german - turkish chamber of trade and commerce, istanbul, 10 february 2015. * 1. * * introduction ladies and gentlemen thank you for inviting me to speak here at the german - turkish chamber of industry and commerce in istanbul. isaac newton once observed that “ we build too many walls and not enough bridges ”. istanbul has for centuries been a bridge between two continents – asia and europe. it seems to me that istanbul is therefore an excellent place to hold this year ’ s g20 meetings. after all, the purpose of these meetings is to build bridges between nations by fostering dialogue and the mutual exchange of information. and this objective is of course also pursued by the german - turkish chamber of trade and commerce. moreover, mutual exchange is essential in a globalised world where countries are no longer isolated islands but part of ...</code> | <code>1.0</code> |\n | <code>for vital issues regarding the agenda. i desire you all success with the materialization and accomplishment of your present and future projects and i conclude by wishing all the foreign guests at this conference a good stay in romania. i am now giving the floor to mr kurt puchinger, coordinator of the eu strategy for the danube region and chair of the first session, to start the conference. thank you. bis central bankers ’ speeches</code> | <code>national bank of romania, as they capture appropriately the opinions of the real sector on some key issues, such as : ( i ) the most pressing problems that firms are facing in their activity ; ( ii ) the investment needs and priorities ; ( iii ) challenges raised by climate change and energy efficiency. the survey shows that firms in romania have become less optimistic regarding investment conditions for the year ahead. while this is a worrying trend, we are rather positive that romania will be able to overcome the main problem identified, the energy costs, in the near future. i would add here that the government has implemented several measures in this respect, such as electricity and natural gas price capping schemes, the compensation of the price of motor fuel / liter and caps on the firewood price. also, we are aware that the war in ukraine and the related sanctions will continue to generate considerable uncertainties and risks to the outlook for economic activity, through possibly...</code> | <code>0.5</code> |\n* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:\n ```json\n {\n \"scale\": 20.0,\n \"similarity_fct\": \"pairwise_cos_sim\"\n }\n ```\n\n### Training Hyperparameters\n#### Non-Default Hyperparameters\n\n- `eval_strategy`: steps\n- `per_device_train_batch_size`: 16\n- `per_device_eval_batch_size`: 16\n- `learning_rate`: 2e-05\n- `num_train_epochs`: 2\n- `warmup_ratio`: 0.1\n- `fp16`: True\n- `batch_sampler`: no_duplicates\n\n#### All Hyperparameters\n<details><summary>Click to expand</summary>\n\n- `overwrite_output_dir`: False\n- `do_predict`: False\n- `eval_strategy`: steps\n- `prediction_loss_only`: True\n- `per_device_train_batch_size`: 16\n- `per_device_eval_batch_size`: 16\n- `per_gpu_train_batch_size`: None\n- `per_gpu_eval_batch_size`: None\n- `gradient_accumulation_steps`: 1\n- `eval_accumulation_steps`: None\n- `torch_empty_cache_steps`: None\n- `learning_rate`: 2e-05\n- `weight_decay`: 0.0\n- `adam_beta1`: 0.9\n- `adam_beta2`: 0.999\n- `adam_epsilon`: 1e-08\n- `max_grad_norm`: 1.0\n- `num_train_epochs`: 2\n- `max_steps`: -1\n- `lr_scheduler_type`: linear\n- `lr_scheduler_kwargs`: {}\n- `warmup_ratio`: 0.1\n- `warmup_steps`: 0\n- `log_level`: passive\n- `log_level_replica`: warning\n- `log_on_each_node`: True\n- `logging_nan_inf_filter`: True\n- `save_safetensors`: True\n- `save_on_each_node`: False\n- `save_only_model`: False\n- `restore_callback_states_from_checkpoint`: False\n- `no_cuda`: False\n- `use_cpu`: False\n- `use_mps_device`: False\n- `seed`: 42\n- `data_seed`: None\n- `jit_mode_eval`: False\n- `use_ipex`: False\n- `bf16`: False\n- `fp16`: True\n- `fp16_opt_level`: O1\n- `half_precision_backend`: auto\n- `bf16_full_eval`: False\n- `fp16_full_eval`: False\n- `tf32`: None\n- `local_rank`: 0\n- `ddp_backend`: None\n- `tpu_num_cores`: None\n- `tpu_metrics_debug`: False\n- `debug`: []\n- `dataloader_drop_last`: False\n- `dataloader_num_workers`: 0\n- `dataloader_prefetch_factor`: None\n- `past_index`: -1\n- `disable_tqdm`: False\n- `remove_unused_columns`: True\n- `label_names`: None\n- `load_best_model_at_end`: False\n- `ignore_data_skip`: False\n- `fsdp`: []\n- `fsdp_min_num_params`: 0\n- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}\n- `fsdp_transformer_layer_cls_to_wrap`: None\n- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}\n- `deepspeed`: None\n- `label_smoothing_factor`: 0.0\n- `optim`: adamw_torch\n- `optim_args`: None\n- `adafactor`: False\n- `group_by_length`: False\n- `length_column_name`: length\n- `ddp_find_unused_parameters`: None\n- `ddp_bucket_cap_mb`: None\n- `ddp_broadcast_buffers`: False\n- `dataloader_pin_memory`: True\n- `dataloader_persistent_workers`: False\n- `skip_memory_metrics`: True\n- `use_legacy_prediction_loop`: False\n- `push_to_hub`: False\n- `resume_from_checkpoint`: None\n- `hub_model_id`: None\n- `hub_strategy`: every_save\n- `hub_private_repo`: None\n- `hub_always_push`: False\n- `gradient_checkpointing`: False\n- `gradient_checkpointing_kwargs`: None\n- `include_inputs_for_metrics`: False\n- `include_for_metrics`: []\n- `eval_do_concat_batches`: True\n- `fp16_backend`: auto\n- `push_to_hub_model_id`: None\n- `push_to_hub_organization`: None\n- `mp_parameters`: \n- `auto_find_batch_size`: False\n- `full_determinism`: False\n- `torchdynamo`: None\n- `ray_scope`: last\n- `ddp_timeout`: 1800\n- `torch_compile`: False\n- `torch_compile_backend`: None\n- `torch_compile_mode`: None\n- `dispatch_batches`: None\n- `split_batches`: None\n- `include_tokens_per_second`: False\n- `include_num_input_tokens_seen`: False\n- `neftune_noise_alpha`: None\n- `optim_target_modules`: None\n- `batch_eval_metrics`: False\n- `eval_on_start`: False\n- `use_liger_kernel`: False\n- `eval_use_gather_object`: False\n- `average_tokens_across_devices`: False\n- `prompts`: None\n- `batch_sampler`: no_duplicates\n- `multi_dataset_batch_sampler`: proportional\n\n</details>\n\n### Training Logs\n| Epoch | Step | Training Loss | Validation Loss | econosentencev2_spearman_cosine |\n|:------:|:----:|:-------------:|:---------------:|:-------------------------------:|\n| 0.0888 | 250 | 4.2245 | - | - | \n| 0.1776 | 500 | 3.6201 | 3.5284 | 0.7812 | \n| 0.2663 | 750 | 3.5093 | - | - | \n| 0.3551 | 1000 | 3.4858 | 3.4938 | 0.8177 | \n| 0.4439 | 1250 | 3.5671 | - | - | \n| 0.5327 | 1500 | 3.3165 | 3.2432 | 0.8302 | \n| 0.6214 | 1750 | 3.2781 | - | - | \n| 0.7102 | 2000 | 3.2901 | 3.1754 | 0.8355 | \n| 0.7990 | 2250 | 3.2053 | - | - | \n| 0.8878 | 2500 | 3.1673 | 3.1109 | 0.8422 | \n| 0.9766 | 2750 | 3.0844 | - | - | \n| 1.0653 | 3000 | 2.7478 | 3.1557 | 0.8483 | \n| 1.1541 | 3250 | 2.7042 | - | - | \n| 1.2429 | 3500 | 2.6291 | 3.1707 | 0.8541 | \n| 1.3317 | 3750 | 2.6193 | - | - | \n| 1.4205 | 4000 | 2.6128 | 3.1136 | 0.8546 | \n| 1.5092 | 4250 | 2.5762 | - | - | \n| 1.5980 | 4500 | 2.5992 | 3.0960 | 0.8579 | \n| 1.6868 | 4750 | 2.4983 | - | - | \n| 1.7756 | 5000 | 2.3061 | 3.1615 | 0.8608 | \n| 1.8643 | 5250 | 2.3459 | - | - | \n| 1.9531 | 5500 | 2.418 | 3.1436 | 0.8617 | \n\n\n### Framework Versions\n- Python: 3.10.12\n- Sentence Transformers: 3.4.1\n- Transformers: 4.49.0\n- PyTorch: 2.1.0+cu118\n- Accelerate: 1.4.0\n- Datasets: 3.3.1\n- Tokenizers: 0.21.0\n\n## Citation\n\nSamuel Chaineau\n\n### BibTeX\n\n#### Sentence Transformers\n```bibtex\n@inproceedings{reimers-2019-sentence-bert,\n title = \"Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks\",\n author = \"Reimers, Nils and Gurevych, Iryna\",\n booktitle = \"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing\",\n month = \"11\",\n year = \"2019\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://arxiv.org/abs/1908.10084\",\n}\n```\n\n#### CoSENTLoss\n```bibtex\n@online{kexuefm-8847,\n title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},\n author={Su Jianlin},\n year={2022},\n month={Jan},\n url={https://kexue.fm/archives/8847},\n}\n```\n\n<!--\n## Glossary\n\n*Clearly define terms in order to be accessible across audiences.*\n-->\n\n<!--\n## Model Card Authors\n\n*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*\n-->\n\n<!--\n## Model Card Contact\n\n*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*\n-->"},"matched_task":{"kind":"list like","value":["TEXT_CLASSIFICATION","SEMANTIC_SIMILARITY"],"string":"[\n \"TEXT_CLASSIFICATION\",\n \"SEMANTIC_SIMILARITY\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["BEAR"],"string":"[\n \"BEAR\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"Non_BioNLP"}}},{"rowIdx":299,"cells":{"id":{"kind":"string","value":"AventIQ-AI/bert-medical-entity-extraction"},"author":{"kind":"string","value":"AventIQ-AI"},"task_category":{"kind":"null"},"tags":{"kind":"list like","value":["safetensors","bert","region:us"],"string":"[\n \"safetensors\",\n \"bert\",\n \"region:us\"\n]","isConversational":false},"created_time":{"kind":"number","value":1740,"string":"1,740"},"last_modified":{"kind":"number","value":1740,"string":"1,740"},"downloads":{"kind":"number","value":72,"string":"72"},"likes":{"kind":"number","value":3,"string":"3"},"README":{"kind":"string","value":"---\n{}\n---\n# Medical Entity Extraction with BERT\n\n## 📌 Overview\nThis repository hosts the quantized version of the `bert-base-cased` model for Medical Entity Extraction using the 'tner/bc5cdr' dataset. The model is specifically designed to recognize entities related to **Disease,Symptoms,Drug**. The model has been optimized for efficient deployment while maintaining high accuracy, making it suitable for resource-constrained environments.\n\n## 🏗 Model Details\n- **Model Architecture**: BERT Base Cased\n- **Task**: Medical Entity Extraction\n- **Dataset**: Hugging Face's `tner/bc5cdr`\n- **Quantization**: Float16\n- **Fine-tuning Framework**: Hugging Face Transformers\n\n---\n## 🚀 Usage\n\n### Installation\n```bash\npip install transformers torch\n```\n\n### Loading the Model\n```python\nfrom transformers import BertTokenizerFast, BertForTokenClassification\nimport torch\n\ndevice = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\nmodel_name = \"AventIQ-AI/bert-medical-entity-extraction\"\nmodel = BertForTokenClassification.from_pretrained(model_name).to(device)\ntokenizer = BertTokenizerFast.from_pretrained(model_name)\n```\n\n### Named Entity Recognition Inference\n```python\nfrom transformers import pipeline\n\nner_pipeline = pipeline(\"ner\", model=model_name, tokenizer=tokenizer)\ntest_sentence = \"An overdose of Ibuprofen can lead to severe gastric issues.\"\nner_results = ner_pipeline(test_sentence)\nlabel_map = {\n \"LABEL_0\": \"O\", # Outside (not an entity)\n \"LABEL_1\": \"Drug\",\n \"LABEL_2\": \"Disease\",\n \"LABEL_3\": \"Symptom\",\n \"LABEL_4\": \"Treatment\"\n}\n\ndef merge_tokens(ner_results):\n merged_entities = []\n current_word = \"\"\n current_label = \"\"\n current_score = 0\n count = 0\n\n for entity in ner_results:\n word = entity[\"word\"]\n label = entity[\"entity\"] # Model's output (e.g., LABEL_1, LABEL_2)\n score = entity[\"score\"]\n\n # Merge subwords\n if word.startswith(\"##\"):\n current_word += word[2:] # Remove '##' and append\n current_score += score\n count += 1\n else:\n if current_word: # Store the previous merged word\n mapped_label = label_map.get(current_label, \"Unknown\")\n merged_entities.append((current_word, mapped_label, current_score / count))\n current_word = word\n current_label = label\n current_score = score\n count = 1\n\n # Add the last word\n if current_word:\n mapped_label = label_map.get(current_label, \"Unknown\")\n merged_entities.append((current_word, mapped_label, current_score / count))\n\n return merged_entities\n\nprint(\"\\n🩺 Medical NER Predictions:\")\nfor word, label, score in merge_tokens(ner_results):\n if label != \"O\": # Skip non-entities\n print(f\"🔹 Entity: {word} | Category: {label} | Score: {score:.4f}\")\n```\n### **🔹 Labeling Scheme (BIO Format)**\n \n- **B-XYZ (Beginning)**: Indicates the beginning of an entity of type XYZ (e.g., B-PER for the beginning of a person’s name).\n- **I-XYZ (Inside)**: Represents subsequent tokens inside an entity (e.g., I-PER for the second part of a person’s name).\n- **O (Outside)**: Denotes tokens that are not part of any named entity.\n\n---\n## 📊 Evaluation Results for Quantized Model\n \n### **🔹 Overall Performance**\n \n- **Accuracy**: **93.27%** ✅\n- **Precision**: **92.31%**\n- **Recall**: **93.27%**\n- **F1 Score**: **92.31%**\n \n---\n \n### **🔹 Performance by Entity Type**\n \n| Entity Type | Precision | Recall | F1 Score | Number of Entities |\n|------------|-----------|--------|----------|--------------------|\n| **Disease** | **91.46%** | **92.07%** | **91.76%** | 3,000 |\n| **Drug** | **71.25%** | **72.83%** | **72.03%** | 1,266 |\n| **Symptom** | **89.83%** | **93.02%** | **91.40%** | 3,524 |\n| **Treatment** | **88.83%** | **92.02%** | **90.40%** | 3,124 |\n\n\n ---\n#### ⏳ **Inference Speed Metrics**\n- **Total Evaluation Time**: 15.89 sec\n- **Samples Processed per Second**: 217.26\n- **Steps per Second**: 27.18\n- **Epochs Completed**: 3\n \n---\n## Fine-Tuning Details\n### Dataset\nThe Hugging Face's `tner/bc5cdr` dataset was used, containing texts and their ner tags.\n\n## 📊 Training Details\n- **Number of epochs**: 3 \n- **Batch size**: 8 \n- **Evaluation strategy**: epoch\n- **Learning Rate**: 2e-5\n\n### ⚡ Quantization\nPost-training quantization was applied using PyTorch's built-in quantization framework to reduce the model size and improve inference efficiency.\n\n---\n## 📂 Repository Structure\n```\n.\n├── model/ # Contains the quantized model files\n├── tokenizer_config/ # Tokenizer configuration and vocabulary files\n├── model.safetensors/ # Quantized Model\n├── README.md # Model documentation\n```\n\n---\n## ⚠️ Limitations\n- The model may not generalize well to domains outside the fine-tuning dataset.\n- Quantization may result in minor accuracy degradation compared to full-precision models.\n\n---\n## 🤝 Contributing\nContributions are welcome! Feel free to open an issue or submit a pull request if you have suggestions or improvements.\n"},"matched_task":{"kind":"list like","value":["NAMED_ENTITY_RECOGNITION"],"string":"[\n \"NAMED_ENTITY_RECOGNITION\"\n]","isConversational":false},"matched_bigbio_names":{"kind":"list like","value":["BC5CDR"],"string":"[\n \"BC5CDR\"\n]","isConversational":false},"is_bionlp":{"kind":"string","value":"BioNLP"}}}],"truncated":false,"partial":false},"paginationData":{"pageIndex":2,"numItemsPerPage":100,"numTotalItems":1838,"offset":200,"length":100}},"jwt":"eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sImlhdCI6MTc3NzQ1MjE1NCwic3ViIjoiL2RhdGFzZXRzL0V1YW55dS9tZXJnZWRfdGFza19kYXRhc2V0X21vZGVsc19CaW9OTFBfcmVzdWx0cyIsImV4cCI6MTc3NzQ1NTc1NCwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.Sal1FGr3d58LPLpSYTgGL4BU_ww62K5bhYDVLFlodIsjM-hhEGderl2YulC7WNthkMIOrWA23WAGrbHs9V6LBQ","displayUrls":true,"splitSizeSummaries":[{"config":"default","split":"train","numRows":1838,"numBytesParquet":48294163}]},"discussionsStats":{"closed":0,"open":1,"total":1},"fullWidth":true,"hasGatedAccess":true,"hasFullAccess":true,"isEmbedded":false,"savedQueries":{"community":[],"user":[]}}">
id stringlengths 11 95 | author stringlengths 3 36 | task_category stringclasses 16
values | tags listlengths 1 4.05k | created_time int64 1.65k 1.74k | last_modified int64 1.62k 1.74k | downloads int64 0 15.6M | likes int64 0 4.86k | README stringlengths 246 1.01M | matched_task listlengths 1 8 | matched_bigbio_names listlengths 1 8 | is_bionlp stringclasses 3
values |
|---|---|---|---|---|---|---|---|---|---|---|---|
croissantllm/base_150k | croissantllm | text2text-generation | [
"transformers",
"pytorch",
"llama",
"text-generation",
"legal",
"code",
"text-generation-inference",
"art",
"text2text-generation",
"fr",
"en",
"dataset:cerebras/SlimPajama-627B",
"dataset:uonlp/CulturaX",
"dataset:pg19",
"dataset:bigcode/starcoderdata",
"license:mit",
"autotrain_com... | 1,704 | 1,706 | 6 | 0 | ---
datasets:
- cerebras/SlimPajama-627B
- uonlp/CulturaX
- pg19
- bigcode/starcoderdata
language:
- fr
- en
license: mit
pipeline_tag: text2text-generation
tags:
- legal
- code
- text-generation-inference
- art
---
# CroissantLLM - Base (150k steps)
This model is part of the CroissantLLM initiative, and corresponds ... | [
"TRANSLATION"
] | [
"CRAFT"
] | Non_BioNLP |
Shashwat13333/bge-base-en-v1.5_v1 | Shashwat13333 | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:150",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-base... | 1,738 | 1,738 | 13 | 0 | ---
base_model: BAAI/bge-base-en-v1.5
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
... | [
"TEXT_CLASSIFICATION"
] | [
"CRAFT"
] | TBD |
RichardErkhov/BSC-LT_-_salamandra-2b-instruct-4bits | RichardErkhov | null | [
"safetensors",
"llama",
"arxiv:2403.14009",
"arxiv:2403.20266",
"arxiv:2101.00027",
"arxiv:2207.00220",
"arxiv:1810.06694",
"arxiv:1911.05507",
"arxiv:1906.03741",
"arxiv:2406.17557",
"arxiv:2402.06619",
"arxiv:1803.09010",
"4-bit",
"bitsandbytes",
"region:us"
] | 1,729 | 1,729 | 4 | 0 | ---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
salamandra-2b-instruct - bnb 4bits
- Model creator: https://huggingface.co/BSC-LT/
- Original model: https://hugg... | [
"QUESTION_ANSWERING",
"TRANSLATION",
"SUMMARIZATION",
"PARAPHRASING"
] | [
"BEAR",
"SCIELO"
] | Non_BioNLP |
michaelfeil/ct2fast-pythia-160m | michaelfeil | null | [
"transformers",
"ctranslate2",
"int8",
"float16",
"pytorch",
"causal-lm",
"pythia",
"en",
"dataset:the_pile",
"arxiv:2101.00027",
"arxiv:2201.07311",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 1,682 | 1,684 | 12 | 1 | ---
datasets:
- the_pile
language:
- en
license: apache-2.0
tags:
- ctranslate2
- int8
- float16
- pytorch
- causal-lm
- pythia
---
# # Fast-Inference with Ctranslate2
Speedup inference while reducing memory by 2x-4x using int8 inference in C++ on CPU or GPU.
quantized version of [EleutherAI/pythia-160m](https://huggi... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
mav23/gpt-neox-20b-GGUF | mav23 | null | [
"gguf",
"pytorch",
"causal-lm",
"en",
"dataset:EleutherAI/pile",
"arxiv:2204.06745",
"arxiv:2101.00027",
"arxiv:2201.07311",
"arxiv:2104.09864",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 1,728 | 1,729 | 278 | 0 | ---
datasets:
- EleutherAI/pile
language:
- en
license: apache-2.0
tags:
- pytorch
- causal-lm
---
GPT-NeoX-20B is a 20 billion parameter autoregressive language model trained
on [the Pile](https://pile.eleuther.ai/) using the [GPT-NeoX
library](https://github.com/EleutherAI/gpt-neox). Its architecture intentionally... | [
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
davanstrien/query-to-dataset-viewer-descriptions | davanstrien | sentence-similarity | [
"sentence-transformers",
"tensorboard",
"safetensors",
"new",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:1141",
"loss:CachedMultipleNegativesRankingLoss",
"custom_code",
"en",
"dataset:davanstrien/query-to-dataset-viewer-descriptions",
"arxiv:1908.10... | 1,726 | 1,726 | 13 | 1 | ---
base_model: Alibaba-NLP/gte-base-en-v1.5
datasets:
- davanstrien/query-to-dataset-viewer-descriptions
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy
- dot_accuracy
- manhattan_accuracy
- euclidean_accuracy
- max_accuracy
pipeline_tag: sentence-similarity
tags:
- se... | [
"NAMED_ENTITY_RECOGNITION",
"TEXT_CLASSIFICATION",
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"BEAR"
] | Non_BioNLP |
EleutherAI/pythia-6.9b-deduped-v0 | EleutherAI | text-generation | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"causal-lm",
"pythia",
"pythia_v0",
"en",
"dataset:EleutherAI/the_pile_deduplicated",
"arxiv:2101.00027",
"arxiv:2201.07311",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"re... | 1,666 | 1,688 | 558 | 20 | ---
datasets:
- EleutherAI/the_pile_deduplicated
language:
- en
license: apache-2.0
tags:
- pytorch
- causal-lm
- pythia
- pythia_v0
---
The *Pythia Scaling Suite* is a collection of models developed to facilitate
interpretability research. It contains two sets of eight models of sizes
70M, 160M, 410M, 1B, 1.4B, 2.8... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
LoneStriker/OpenBioLLM-Llama3-70B-6.0bpw-h6-exl2 | LoneStriker | text-generation | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-3",
"Mixtral",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"distillation",
"conversational",
"en",
"arxiv:2305.18290",
"arxiv:2303.13375",
"arxiv:2212.13138",
"arxiv:2305.09617",
"arxiv:2402.07023",
"b... | 1,714 | 1,714 | 8 | 0 | ---
base_model: meta-llama/Meta-Llama-3-70B-Instruct
language:
- en
license: llama3
tags:
- llama-3
- llama
- Mixtral
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- distillation
widget:
- example_title: OpenBioLLM-70B
messages:
- role: system
content: You are an expert and experienced from the healthcare ... | [
"QUESTION_ANSWERING"
] | [
"MEDQA",
"PUBMEDQA"
] | BioNLP |
HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 | HIT-TMG | sentence-similarity | [
"sentence-transformers",
"safetensors",
"qwen2",
"feature-extraction",
"sentence-similarity",
"mteb",
"arxiv:2501.01028",
"license:mit",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,729 | 1,741 | 36,178 | 32 | ---
license: mit
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
model-index:
- name: KaLM-Embedding
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en-ext)
type: mteb/amazon_counter... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
maastrichtlawtech/distilcamembert-lleqa | maastrichtlawtech | sentence-similarity | [
"sentence-transformers",
"pytorch",
"safetensors",
"camembert",
"feature-extraction",
"sentence-similarity",
"fr",
"dataset:maastrichtlawtech/lleqa",
"arxiv:2309.17050",
"license:apache-2.0",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,695 | 1,730 | 48 | 3 | ---
datasets:
- maastrichtlawtech/lleqa
language: fr
library_name: sentence-transformers
license: apache-2.0
metrics:
- recall
pipeline_tag: sentence-similarity
tags:
- feature-extraction
- sentence-similarity
inference: true
widget:
- source_sentence: Je reçois des confidences liées à mon emploi. Qu'est-ce que je
... | [
"QUESTION_ANSWERING"
] | [
"CAS"
] | Non_BioNLP |
am-azadi/KaLM-embedding-multilingual-mini-v1_Fine_Tuned_1e | am-azadi | sentence-similarity | [
"sentence-transformers",
"safetensors",
"qwen2",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:21769",
"loss:MultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:HIT-TMG/KaLM-embedding-multilingual-mini-v1",
"base_model:finet... | 1,740 | 1,740 | 16 | 0 | ---
base_model: HIT-TMG/KaLM-embedding-multilingual-mini-v1
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:21769
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: 'Bloomin... | [
"TEXT_CLASSIFICATION"
] | [
"PCR"
] | Non_BioNLP |
markaw/NV-Embed-v2 | markaw | feature-extraction | [
"transformers",
"safetensors",
"nvembed",
"feature-extraction",
"mteb",
"custom_code",
"en",
"arxiv:2405.17428",
"arxiv:2407.15831",
"license:cc-by-nc-4.0",
"model-index",
"endpoints_compatible",
"region:us"
] | 1,727 | 1,727 | 15 | 0 | ---
language:
- en
library_name: transformers
license: cc-by-nc-4.0
tags:
- mteb
model-index:
- name: NV-Embed-v2
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
config: en
split: tes... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
TitanML/jina-v2-base-en-embed | TitanML | feature-extraction | [
"sentence-transformers",
"pytorch",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"custom_code",
"en",
"dataset:allenai/c4",
"arxiv:2108.12409",
"arxiv:2310.19923",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
... | 1,713 | 1,713 | 26 | 0 | ---
datasets:
- allenai/c4
language: en
license: apache-2.0
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
inference: false
model-index:
- name: jina-embedding-b-en-v2
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
RichardErkhov/EleutherAI_-_pythia-2.8b-deduped-v0-gguf | RichardErkhov | null | [
"gguf",
"arxiv:2101.00027",
"arxiv:2201.07311",
"endpoints_compatible",
"region:us"
] | 1,730 | 1,730 | 87 | 0 | ---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
pythia-2.8b-deduped-v0 - GGUF
- Model creator: https://huggingface.co/EleutherAI/
- Original model: https://huggi... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
tomaarsen/mpnet-base-natural-questions-icl | tomaarsen | sentence-similarity | [
"sentence-transformers",
"safetensors",
"mpnet",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:100231",
"loss:ImprovedContrastiveLoss",
"en",
"dataset:sentence-transformers/natural-questions",
"arxiv:1908.10084",
"base_model:microsoft/mpnet-base",
"base... | 1,719 | 1,720 | 4 | 0 | ---
base_model: microsoft/mpnet-base
datasets:
- sentence-transformers/natural-questions
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_prec... | [
"TEXT_CLASSIFICATION"
] | [
"MEDAL"
] | Non_BioNLP |
Gopal2002/Material_Receipt_Report_ZEON | Gopal2002 | text-classification | [
"setfit",
"safetensors",
"bert",
"sentence-transformers",
"text-classification",
"generated_from_setfit_trainer",
"arxiv:2209.11055",
"base_model:BAAI/bge-small-en-v1.5",
"base_model:finetune:BAAI/bge-small-en-v1.5",
"model-index",
"region:us"
] | 1,705 | 1,705 | 4 | 0 | ---
base_model: BAAI/bge-small-en-v1.5
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: "* 04 Hindalco Industries Ltd\nHirkaud Smelter Stores\n\n \n\n* Service Recei\
\ ot\nBUYER _ ... | [
"TEXT_CLASSIFICATION"
] | [
"CAS",
"CPI"
] | Non_BioNLP |
lixsh6/XLM-3B5-embedding | lixsh6 | null | [
"mteb",
"model-index",
"region:us"
] | 1,690 | 1,690 | 0 | 0 | ---
tags:
- mteb
model-index:
- name: xlm3b5_step3len260_b128g8_lr1e-5
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
Muennighoff/SGPT-5.8B-weightedmean-msmarco-specb-bitfit | Muennighoff | sentence-similarity | [
"sentence-transformers",
"pytorch",
"gptj",
"feature-extraction",
"sentence-similarity",
"mteb",
"arxiv:2202.08904",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 1,646 | 1,679 | 83 | 23 | ---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
model-index:
- name: SGPT-5.8B-weightedmean-msmarco-specb-bitfit
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/am... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
RichardErkhov/dunzhang_-_stella_en_1.5B_v5-4bits | RichardErkhov | null | [
"safetensors",
"qwen2",
"custom_code",
"arxiv:2205.13147",
"4-bit",
"bitsandbytes",
"region:us"
] | 1,730 | 1,730 | 4 | 0 | ---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
stella_en_1.5B_v5 - bnb 4bits
- Model creator: https://huggingface.co/dunzhang/
- Original model: https://hugging... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
adriansanz/sitges2608bai-4ep | adriansanz | sentence-similarity | [
"sentence-transformers",
"safetensors",
"xlm-roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:4173",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-m3",... | 1,724 | 1,724 | 4 | 0 | ---
base_model: BAAI/bge-m3
datasets: []
language: []
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
-... | [
"TEXT_CLASSIFICATION"
] | [
"CAS"
] | Non_BioNLP |
BAAI/bge-large-zh | BAAI | feature-extraction | [
"transformers",
"pytorch",
"safetensors",
"bert",
"feature-extraction",
"zh",
"arxiv:2310.07554",
"arxiv:2309.07597",
"license:mit",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,690 | 1,697 | 21,308 | 320 | ---
language:
- zh
license: mit
---
**Recommend switching to newest [BAAI/bge-large-zh-v1.5](https://huggingface.co/BAAI/bge-large-zh-v1.5), which has more reasonable similarity distribution and same method of usage.**
<h1 align="center">FlagEmbedding</h1>
<h4 align="center">
<p>
<a href=#model-list>Mo... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BEAR"
] | Non_BioNLP |
RomainDarous/large_directTwoEpoch_meanPooling_mistranslationModel | RomainDarous | sentence-similarity | [
"sentence-transformers",
"safetensors",
"xlm-roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:4460010",
"loss:CoSENTLoss",
"dataset:RomainDarous/corrupted_os_by_language",
"arxiv:1908.10084",
"base_model:RomainDarous/large_directOneEpoch_meanPooling_... | 1,740 | 1,740 | 22 | 0 | ---
base_model: RomainDarous/large_directOneEpoch_meanPooling_mistranslationModel
datasets:
- RomainDarous/corrupted_os_by_language
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
-... | [
"TEXT_CLASSIFICATION",
"SEMANTIC_SIMILARITY",
"TRANSLATION"
] | [
"CAS"
] | Non_BioNLP |
BAAI/bge-large-zh-noinstruct | BAAI | feature-extraction | [
"transformers",
"pytorch",
"bert",
"feature-extraction",
"zh",
"arxiv:2310.07554",
"arxiv:2309.07597",
"license:mit",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,690 | 1,697 | 69 | 11 | ---
language:
- zh
license: mit
---
**Recommend switching to newest [BAAI/bge-large-zh-v1.5](https://huggingface.co/BAAI/bge-large-zh-v1.5) , which has more reasonable similarity distribution and same method of usage.**
<h1 align="center">FlagEmbedding</h1>
<h4 align="center">
<p>
<a href=#model-list>M... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BEAR"
] | Non_BioNLP |
udrearobert999/multi-qa-mpnet-base-cos-v1-test | udrearobert999 | text-classification | [
"setfit",
"safetensors",
"mpnet",
"sentence-transformers",
"text-classification",
"generated_from_setfit_trainer",
"arxiv:2209.11055",
"base_model:sentence-transformers/multi-qa-mpnet-base-cos-v1",
"base_model:finetune:sentence-transformers/multi-qa-mpnet-base-cos-v1",
"model-index",
"region:us"... | 1,714 | 1,714 | 5 | 2 | ---
base_model: sentence-transformers/multi-qa-mpnet-base-cos-v1
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: in durankulak near varna is another important example other signs of ea... | [
"TEXT_CLASSIFICATION",
"TRANSLATION"
] | [
"PCR"
] | Non_BioNLP |
nold/CroissantLLMChat-v0.1-GGUF | nold | text2text-generation | [
"gguf",
"legal",
"code",
"text-generation-inference",
"art",
"text2text-generation",
"fr",
"en",
"dataset:croissantllm/croissant_dataset",
"dataset:croissantllm/CroissantLLM-2201-sft",
"dataset:cerebras/SlimPajama-627B",
"dataset:uonlp/CulturaX",
"dataset:pg19",
"dataset:bigcode/starcoderd... | 1,707 | 1,708 | 80 | 0 | ---
datasets:
- croissantllm/croissant_dataset
- croissantllm/CroissantLLM-2201-sft
- cerebras/SlimPajama-627B
- uonlp/CulturaX
- pg19
- bigcode/starcoderdata
language:
- fr
- en
license: mit
pipeline_tag: text2text-generation
tags:
- legal
- code
- text-generation-inference
- art
---
# CroissantLLMChat (190k steps + ... | [
"TRANSLATION"
] | [
"CRAFT"
] | Non_BioNLP |
juanpablomesa/bge-base-financial-matryoshka | juanpablomesa | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:9600",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-bas... | 1,719 | 1,719 | 47 | 0 | ---
base_model: BAAI/bge-base-en-v1.5
datasets: []
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cos... | [
"TEXT_CLASSIFICATION"
] | [
"MEDAL"
] | Non_BioNLP |
svorwerk/setfit-fine-tuned-demo-class_hpo | svorwerk | text-classification | [
"setfit",
"safetensors",
"mpnet",
"sentence-transformers",
"text-classification",
"generated_from_setfit_trainer",
"arxiv:2209.11055",
"base_model:sentence-transformers/all-mpnet-base-v2",
"base_model:finetune:sentence-transformers/all-mpnet-base-v2",
"region:us"
] | 1,706 | 1,706 | 5 | 0 | ---
base_model: sentence-transformers/all-mpnet-base-v2
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 'Acquisition Id: ALOG; Ancotel; Asia Tone; Bit-Isle; IXEurope; Infomart; Itconic... | [
"TEXT_CLASSIFICATION",
"TRANSLATION"
] | [
"CHIA",
"MIRNA"
] | Non_BioNLP |
knguyennguyen/mpnet_jacket4k_adjustedv3 | knguyennguyen | sentence-similarity | [
"sentence-transformers",
"safetensors",
"mpnet",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:11097",
"loss:MultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:sentence-transformers/all-mpnet-base-v2",
"base_model:finetune:... | 1,737 | 1,737 | 9 | 0 | ---
base_model: sentence-transformers/all-mpnet-base-v2
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:11097
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: a jacket for... | [
"TEXT_CLASSIFICATION"
] | [
"BEAR"
] | Non_BioNLP |
tessimago/bge-large-repmus-matryoshka | tessimago | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:1024",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-lar... | 1,726 | 1,726 | 9 | 0 | ---
base_model: BAAI/bge-large-en-v1.5
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3... | [
"TEXT_CLASSIFICATION"
] | [
"CRAFT"
] | Non_BioNLP |
delwinn/stella_en_1.5B_v5 | delwinn | feature-extraction | [
"sentence-transformers",
"pytorch",
"safetensors",
"qwen2",
"text-generation",
"mteb",
"transformers",
"sentence-similarity",
"feature-extraction",
"custom_code",
"arxiv:2205.13147",
"license:mit",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"text-embeddings-inf... | 1,731 | 1,731 | 13 | 0 | ---
license: mit
pipeline_tag: feature-extraction
tags:
- mteb
- sentence-transformers
- transformers
- sentence-similarity
model-index:
- name: stella_en_1.5B_v5
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
Mihaiii/test25 | Mihaiii | sentence-similarity | [
"sentence-transformers",
"onnx",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"bge",
"mteb",
"mergekit",
"merge",
"base_model:Mihaiii/Wartortle",
"base_model:merge:Mihaiii/Wartortle",
"base_model:TaylorAI/bge-micro-v2",
"base_model:merge:TaylorAI/bge-micro-v2",
"li... | 1,714 | 1,714 | 11 | 0 | ---
base_model:
- Mihaiii/Wartortle
- TaylorAI/bge-micro-v2
library_name: sentence-transformers
license: mit
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- bge
- mteb
- mergekit
- merge
model-index:
- name: Giratina
results:
- task:
type: Classific... | [
"SUMMARIZATION"
] | [
"BIOSSES"
] | BioNLP |
tner/deberta-v3-large-bc5cdr | tner | token-classification | [
"transformers",
"pytorch",
"deberta-v2",
"token-classification",
"dataset:tner/bc5cdr",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 1,660 | 1,664 | 20 | 0 | ---
datasets:
- tner/bc5cdr
metrics:
- f1
- precision
- recall
pipeline_tag: token-classification
widget:
- text: Jacob Collier is a Grammy awarded artist from England.
example_title: NER Example 1
model-index:
- name: tner/deberta-v3-large-bc5cdr
results:
- task:
type: token-classification
name: Toke... | [
"NAMED_ENTITY_RECOGNITION"
] | [
"BC5CDR"
] | BioNLP |
MediaTek-Research/Breexe-8x7B-Instruct-v0_1 | MediaTek-Research | text-generation | [
"transformers",
"pytorch",
"mixtral",
"text-generation",
"conversational",
"en",
"zh",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 1,707 | 1,722 | 167 | 55 | ---
language:
- en
- zh
license: apache-2.0
pipeline_tag: text-generation
extra_gated_prompt: The model weights are only available for the partners to download
now.
extra_gated_fields:
Name: text
Company: text
Title: text
Contact Email: text
---
# Breexe-8x7B-Instruct-v0_1
Breexe-8x7B is a language model f... | [
"SUMMARIZATION"
] | [
"BEAR"
] | Non_BioNLP |
RichardErkhov/EleutherAI_-_pythia-2.8b-v0-gguf | RichardErkhov | null | [
"gguf",
"arxiv:2101.00027",
"arxiv:2201.07311",
"endpoints_compatible",
"region:us"
] | 1,730 | 1,730 | 73 | 0 | ---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
pythia-2.8b-v0 - GGUF
- Model creator: https://huggingface.co/EleutherAI/
- Original model: https://huggingface.c... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
avsolatorio/all-MiniLM-L6-v2-MEDI-MTEB-triplet-final | avsolatorio | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:1943715",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:sentence-transformers/all-MiniLM-L6-v2",
"base_model:f... | 1,720 | 1,720 | 51 | 1 | ---
base_model: sentence-transformers/all-MiniLM-L6-v2
datasets: []
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy
- dot_accuracy
- manhattan_accuracy
- euclidean_accuracy
- max_accuracy
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similar... | [
"TEXT_CLASSIFICATION",
"SUMMARIZATION",
"PARAPHRASING"
] | [
"PUBMEDQA",
"SCIFACT",
"SCIQ",
"SCITAIL"
] | Non_BioNLP |
ISOISS/jina-embeddings-v3-tei | ISOISS | feature-extraction | [
"transformers",
"pytorch",
"onnx",
"safetensors",
"xlm-roberta",
"feature-extraction",
"sentence-similarity",
"mteb",
"sentence-transformers",
"custom_code",
"multilingual",
"af",
"am",
"ar",
"as",
"az",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"cs",
"cy",
"da",
"de",
... | 1,731 | 1,731 | 9 | 0 | ---
language:
- multilingual
- af
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- en
- eo
- es
- et
- eu
- fa
- fi
- fr
- fy
- ga
- gd
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- id
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
SepKeyPro/bge-base-en-trivia-anchor-positive | SepKeyPro | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:9000",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:BAAI/bge-base-en-v1.5",
"base_model:finetune:BAAI/bge-bas... | 1,719 | 1,719 | 9 | 0 | ---
base_model: BAAI/bge-base-en-v1.5
datasets: []
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cos... | [
"TEXT_CLASSIFICATION",
"TRANSLATION"
] | [
"MEDAL"
] | Non_BioNLP |
fine-tuned/SciFact-32000-384-gpt-4o-2024-05-13-66747460 | fine-tuned | feature-extraction | [
"sentence-transformers",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"en",
"dataset:fine-tuned/SciFact-32000-384-gpt-4o-2024-05-13-66747460",
"dataset:allenai/c4",
"license:apache-2.0",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible... | 1,717 | 1,717 | 6 | 0 | ---
datasets:
- fine-tuned/SciFact-32000-384-gpt-4o-2024-05-13-66747460
- allenai/c4
language:
- en
- en
license: apache-2.0
pipeline_tag: feature-extraction
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
---
This model is a fine-tuned version of [**BAAI/bge-large-en-v1.5**](https://hug... | [
"TEXT_CLASSIFICATION"
] | [
"SCIFACT"
] | Non_BioNLP |
RichardErkhov/M4-ai_-_tau-0.5B-awq | RichardErkhov | null | [
"safetensors",
"qwen2",
"4-bit",
"awq",
"region:us"
] | 1,732 | 1,732 | 4 | 0 | ---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
tau-0.5B - AWQ
- Model creator: https://huggingface.co/M4-ai/
- Original model: https://huggingface.co/M4-ai/tau-... | [
"QUESTION_ANSWERING",
"SUMMARIZATION"
] | [
"MEDQA"
] | Non_BioNLP |
adriansanz/SITGES-bge-FT2 | adriansanz | sentence-similarity | [
"sentence-transformers",
"safetensors",
"xlm-roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:237",
"loss:BatchAllTripletLoss",
"arxiv:1908.10084",
"arxiv:1703.07737",
"base_model:BAAI/bge-m3",
"base_model:finetune:BAAI/bge-m3",
"autotrain_compat... | 1,723 | 1,723 | 5 | 0 | ---
base_model: BAAI/bge-m3
datasets: []
language: []
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:237
- loss:BatchAllTripletLoss
widget:
- source_sentence: 'El Viver dels Avis de Si... | [
"TEXT_CLASSIFICATION"
] | [
"CAS"
] | Non_BioNLP |
ginkgogo/setfit-absa-bge-small-en-v1.5-restaurants-aspect | ginkgogo | text-classification | [
"setfit",
"safetensors",
"bert",
"absa",
"sentence-transformers",
"text-classification",
"generated_from_setfit_trainer",
"arxiv:2209.11055",
"base_model:sentence-transformers/all-MiniLM-L6-v2",
"base_model:finetune:sentence-transformers/all-MiniLM-L6-v2",
"model-index",
"region:us"
] | 1,711 | 1,711 | 4 | 0 | ---
base_model: sentence-transformers/all-MiniLM-L6-v2
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- absa
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: variety:I'm not sure what what I would do if I'd never discovered Nikka, s... | [
"TEXT_CLASSIFICATION"
] | [
"CHIA"
] | Non_BioNLP |
vectoriseai/multilingual-e5-large | vectoriseai | feature-extraction | [
"sentence-transformers",
"pytorch",
"onnx",
"safetensors",
"xlm-roberta",
"mteb",
"Sentence Transformers",
"sentence-similarity",
"feature-extraction",
"multilingual",
"af",
"am",
"ar",
"as",
"az",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"cs",
"cy",
"da",
"de",
"el",
... | 1,696 | 1,696 | 14 | 0 | ---
language:
- multilingual
- af
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- en
- eo
- es
- et
- eu
- fa
- fi
- fr
- fy
- ga
- gd
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- id
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
- lo
- lt
- lv
- mg
- mk
- ml
- mn
- mr
- ms
- my
-... | [
"SEMANTIC_SIMILARITY",
"TRANSLATION",
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
tanbinh2210/mlm_finetuned_3_phobert | tanbinh2210 | sentence-similarity | [
"sentence-transformers",
"safetensors",
"roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:357018",
"loss:MultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:tanbinh2210/mlm_finetuned_2_phobert",
"base_model:finetune:t... | 1,732 | 1,732 | 7 | 0 | ---
base_model: tanbinh2210/mlm_finetuned_2_phobert
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:357018
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: đánh_giá phẩm_c... | [
"TEXT_CLASSIFICATION"
] | [
"CHIA"
] | Non_BioNLP |
jinaai/jina-embeddings-v3 | jinaai | feature-extraction | [
"transformers",
"pytorch",
"onnx",
"safetensors",
"feature-extraction",
"sentence-similarity",
"mteb",
"sentence-transformers",
"custom_code",
"multilingual",
"af",
"am",
"ar",
"as",
"az",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"cs",
"cy",
"da",
"de",
"el",
"en",
"... | 1,725 | 1,740 | 1,734,211 | 828 | ---
language:
- multilingual
- af
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- en
- eo
- es
- et
- eu
- fa
- fi
- fr
- fy
- ga
- gd
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- id
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
- lo
- lt
- lv
- mg
- mk
- ml
- mn
- mr
- ms
- my
-... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
meandyou200175/phobert_ContrastiveLoss | meandyou200175 | sentence-similarity | [
"sentence-transformers",
"safetensors",
"roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:87608",
"loss:ContrastiveLoss",
"arxiv:1908.10084",
"base_model:vinai/phobert-base-v2",
"base_model:finetune:vinai/phobert-base-v2",
"model-index",
"autotra... | 1,730 | 1,730 | 8 | 0 | ---
base_model: vinai/phobert-base-v2
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@2
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_accuracy@100
- cosine_precision@1
- cosine_precision@2
- cosine_precision@5
- cosine_precision@10
- cosine_precision@100
- cosine_recall@1
- cosine... | [
"TEXT_CLASSIFICATION"
] | [
"CHIA"
] | BioNLP |
twadada/nmc-cls-nobiv | twadada | null | [
"mteb",
"model-index",
"region:us"
] | 1,725 | 1,725 | 0 | 0 | ---
tags:
- mteb
model-index:
- name: nomic_classification_nobivec
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: None
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
- type: a... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
twadada/nmc-prmt-and-emb | twadada | null | [
"mteb",
"model-index",
"region:us"
] | 1,725 | 1,725 | 0 | 0 | ---
tags:
- mteb
model-index:
- name: nomic_classification_prompt_domain_sample
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: None
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
jinaai/jina-embedding-l-en-v1 | jinaai | sentence-similarity | [
"sentence-transformers",
"pytorch",
"t5",
"finetuner",
"mteb",
"feature-extraction",
"sentence-similarity",
"custom_code",
"en",
"dataset:jinaai/negation-dataset",
"arxiv:2307.11224",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 1,688 | 1,736 | 561 | 24 | ---
datasets:
- jinaai/negation-dataset
language: en
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- finetuner
- mteb
- sentence-transformers
- feature-extraction
- sentence-similarity
model-index:
- name: jina-triplets-large
results:
- task:
type: Classification
dataset:
name: MTEB Am... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"LINNAEUS",
"SCIFACT"
] | Non_BioNLP |
TaylorAI/gte-tiny | TaylorAI | sentence-similarity | [
"sentence-transformers",
"pytorch",
"onnx",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"transformers",
"mteb",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,696 | 1,696 | 59,999 | 136 | ---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
- mteb
model-index:
- name: gte_tiny
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
Nizzouuu/gte-Qwen2-7B-instruct-Q6_K-GGUF | Nizzouuu | sentence-similarity | [
"sentence-transformers",
"gguf",
"mteb",
"transformers",
"Qwen2",
"sentence-similarity",
"llama-cpp",
"gguf-my-repo",
"base_model:Alibaba-NLP/gte-Qwen2-7B-instruct",
"base_model:quantized:Alibaba-NLP/gte-Qwen2-7B-instruct",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endp... | 1,738 | 1,738 | 7 | 0 | ---
base_model: Alibaba-NLP/gte-Qwen2-7B-instruct
license: apache-2.0
tags:
- mteb
- sentence-transformers
- transformers
- Qwen2
- sentence-similarity
- llama-cpp
- gguf-my-repo
model-index:
- name: gte-qwen2-7B-instruct
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactual... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
SeaLLMs/SeaLLMs-v3-1.5B-Chat | SeaLLMs | text-generation | [
"transformers",
"safetensors",
"qwen2",
"text-generation",
"sea",
"multilingual",
"conversational",
"en",
"zh",
"id",
"vi",
"th",
"ms",
"tl",
"ta",
"jv",
"arxiv:2407.19672",
"arxiv:2306.05179",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints... | 1,721 | 1,722 | 1,507 | 12 | ---
language:
- en
- zh
- id
- vi
- th
- ms
- tl
- ta
- jv
license: other
license_name: seallms
license_link: https://huggingface.co/SeaLLMs/SeaLLM-13B-Chat/blob/main/LICENSE
tags:
- sea
- multilingual
---
# *SeaLLMs-v3* - Large Language Models for Southeast Asia
<p align="center">
<a href="https://damo-nlp-sg.githu... | [
"TRANSLATION"
] | [
"CHIA"
] | Non_BioNLP |
jukofyork/Dark-Miqu-70B | jukofyork | text-generation | [
"transformers",
"safetensors",
"llama",
"text-generation",
"mergekit",
"merge",
"arxiv:2403.19522",
"base_model:152334H/miqu-1-70b-sf",
"base_model:merge:152334H/miqu-1-70b-sf",
"base_model:Sao10K/Euryale-1.3-L2-70B",
"base_model:merge:Sao10K/Euryale-1.3-L2-70B",
"base_model:Sao10K/WinterGodde... | 1,714 | 1,715 | 72 | 30 | ---
base_model:
- 152334H/miqu-1-70b-sf
- sophosympatheia/Midnight-Rose-70B-v2.0.3
- Sao10K/Euryale-1.3-L2-70B
- Sao10K/WinterGoddess-1.4x-70B-L2
library_name: transformers
license: other
tags:
- mergekit
- merge
---
A "dark" creative writing model with 32k context. Based off [miqu-1-7... | [
"TRANSLATION"
] | [
"BEAR"
] | Non_BioNLP |
croissantllm/base_175k | croissantllm | text2text-generation | [
"transformers",
"pytorch",
"llama",
"text-generation",
"legal",
"code",
"text-generation-inference",
"art",
"text2text-generation",
"fr",
"en",
"dataset:cerebras/SlimPajama-627B",
"dataset:uonlp/CulturaX",
"dataset:pg19",
"dataset:bigcode/starcoderdata",
"license:mit",
"autotrain_com... | 1,705 | 1,706 | 6 | 0 | ---
datasets:
- cerebras/SlimPajama-627B
- uonlp/CulturaX
- pg19
- bigcode/starcoderdata
language:
- fr
- en
license: mit
pipeline_tag: text2text-generation
tags:
- legal
- code
- text-generation-inference
- art
---
# CroissantLLM - Base (175k steps)
This model is part of the CroissantLLM initiative, and corresponds ... | [
"TRANSLATION"
] | [
"CRAFT"
] | Non_BioNLP |
Teradata/bge-small-en-v1.5 | Teradata | feature-extraction | [
"onnx",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"teradata",
"en",
"license:mit",
"model-index",
"region:us"
] | 1,739 | 1,741 | 28 | 0 | ---
language:
- en
license: mit
tags:
- feature-extraction
- sentence-similarity
- mteb
- onnx
- teradata
model-index:
- name: bge-small-en-v1.5
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
config: en
... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
fine-tuned/SciFact-512-192-gpt-4o-2024-05-13-65992666 | fine-tuned | feature-extraction | [
"sentence-transformers",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"en",
"dataset:fine-tuned/SciFact-512-192-gpt-4o-2024-05-13-65992666",
"dataset:allenai/c4",
"license:apache-2.0",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",... | 1,716 | 1,716 | 6 | 0 | ---
datasets:
- fine-tuned/SciFact-512-192-gpt-4o-2024-05-13-65992666
- allenai/c4
language:
- en
- en
license: apache-2.0
pipeline_tag: feature-extraction
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
---
This model is a fine-tuned version of [**BAAI/bge-large-en-v1.5**](https://huggi... | [
"TEXT_CLASSIFICATION"
] | [
"SCIFACT"
] | Non_BioNLP |
legalvn/paraphrase-multilingual-MiniLM-L12-v2-vn-170000 | legalvn | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:651725",
"loss:SoftmaxLoss",
"arxiv:1908.10084",
"base_model:sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
"base_model:finetune:sentence-transforme... | 1,733 | 1,733 | 6 | 0 | ---
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:651725
- loss:SoftmaxLoss
widget:
- source_sentence: Nguyên t... | [
"TEXT_CLASSIFICATION"
] | [
"PCR"
] | Non_BioNLP |
bobox/DeBERTaV3-small-GeneralSentenceTransformer-v2 | bobox | sentence-similarity | [
"sentence-transformers",
"pytorch",
"deberta-v2",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:96781",
"loss:MultipleNegativesRankingLoss",
"loss:AnglELoss",
"loss:GISTEmbedLoss",
"loss:OnlineContrastiveLoss",
"loss:MultipleNegativesSymmetricRankingLoss"... | 1,718 | 1,718 | 13 | 0 | ---
base_model: microsoft/deberta-v3-small
datasets:
- sentence-transformers/all-nli
- sentence-transformers/stsb
- tals/vitaminc
- nyu-mll/glue
- allenai/scitail
- sentence-transformers/xsum
- sentence-transformers/sentence-compression
language:
- en
library_name: sentence-transformers
pipeline_tag: sentence-similarit... | [
"TEXT_CLASSIFICATION"
] | [
"SCITAIL"
] | Non_BioNLP |
EleutherAI/pythia-1b-deduped | EleutherAI | text-generation | [
"transformers",
"pytorch",
"safetensors",
"gpt_neox",
"text-generation",
"causal-lm",
"pythia",
"en",
"dataset:EleutherAI/the_pile_deduplicated",
"arxiv:2304.01373",
"arxiv:2101.00027",
"arxiv:2201.07311",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endp... | 1,676 | 1,689 | 20,123 | 19 | ---
datasets:
- EleutherAI/the_pile_deduplicated
language:
- en
license: apache-2.0
tags:
- pytorch
- causal-lm
- pythia
---
The *Pythia Scaling Suite* is a collection of models developed to facilitate
interpretability research [(see paper)](https://arxiv.org/pdf/2304.01373.pdf).
It contains two sets of eight models... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
corto-ai/nomic-embed-text-v1 | corto-ai | sentence-similarity | [
"sentence-transformers",
"pytorch",
"onnx",
"safetensors",
"nomic_bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"transformers",
"transformers.js",
"custom_code",
"en",
"arxiv:2402.01613",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-embeddings-infe... | 1,714 | 1,714 | 740 | 2 | ---
language:
- en
library_name: sentence-transformers
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- feature-extraction
- sentence-similarity
- mteb
- transformers
- transformers.js
model-index:
- name: epoch_0_model
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCou... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
blockblockblock/Dark-Miqu-70B-bpw4-exl2 | blockblockblock | text-generation | [
"transformers",
"safetensors",
"llama",
"text-generation",
"arxiv:2403.19522",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"exl2",
"region:us"
] | 1,715 | 1,715 | 14 | 2 | ---
license: other
---
***NOTE***: *For a full range of GGUF quants kindly provided by @mradermacher: [Static](https://huggingface.co/mradermacher/Dark-Miqu-70B-GGUF) and [IMatrix](https://huggingface.co/mradermacher/Dark-Miqu-70B-i1-GGUF).*
A "dark" creative writing model with 32k co... | [
"TRANSLATION"
] | [
"BEAR"
] | Non_BioNLP |
Haon-Chen/speed-embedding-7b-instruct | Haon-Chen | feature-extraction | [
"transformers",
"safetensors",
"mistral",
"feature-extraction",
"mteb",
"en",
"arxiv:2410.18634",
"arxiv:2104.08663",
"arxiv:2210.07316",
"license:mit",
"model-index",
"text-generation-inference",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,730 | 1,730 | 169 | 5 | ---
language:
- en
license: mit
tags:
- mteb
- transformers
model-index:
- name: speed-embedding-7b-instruct
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
config: en
split: test
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
deepfile/multilingual-e5-small-onnx-qint8 | deepfile | sentence-similarity | [
"sentence-transformers",
"onnx",
"safetensors",
"bert",
"mteb",
"Sentence Transformers",
"sentence-similarity",
"multilingual",
"af",
"am",
"ar",
"as",
"az",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa"... | 1,731 | 1,731 | 10 | 1 | ---
language:
- multilingual
- af
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- en
- eo
- es
- et
- eu
- fa
- fi
- fr
- fy
- ga
- gd
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- id
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
- lo
- lt
- lv
- mg
- mk
- ml
- mn
- mr
- ms
- my
-... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
hcy5561/distilroberta-base-sentence-transformer-snli | hcy5561 | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:494430",
"loss:SoftmaxLoss",
"en",
"dataset:stanfordnlp/snli",
"arxiv:1908.10084",
"base_model:google-bert/bert-base-uncased",
"base_model:finetune:google-bert... | 1,724 | 1,724 | 12 | 0 | ---
base_model: google-bert/bert-base-uncased
datasets:
- stanfordnlp/snli
language:
- en
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:494430
- loss:SoftmaxLoss
widget... | [
"TEXT_CLASSIFICATION"
] | [
"BEAR"
] | Non_BioNLP |
pszemraj/long-t5-tglobal-base-16384-booksum-V11-big_patent-V2 | pszemraj | summarization | [
"transformers",
"pytorch",
"safetensors",
"longt5",
"text2text-generation",
"summarization",
"summary",
"booksum",
"long-document",
"long-form",
"dataset:kmfoda/booksum",
"dataset:big_patent",
"license:apache-2.0",
"license:bsd-3-clause",
"model-index",
"autotrain_compatible",
"endpo... | 1,659 | 1,692 | 87 | 2 | ---
datasets:
- kmfoda/booksum
- big_patent
license:
- apache-2.0
- bsd-3-clause
metrics:
- rouge
tags:
- summarization
- summary
- booksum
- long-document
- long-form
widget:
- text: large earthquakes along a given fault segment do not occur at random intervals
because it takes time to accumulate the strain energy... | [
"QUESTION_ANSWERING",
"SUMMARIZATION"
] | [
"BEAR"
] | Non_BioNLP |
MilosKosRad/BioNER | MilosKosRad | token-classification | [
"transformers",
"pytorch",
"bert",
"token-classification",
"chemistry",
"biology",
"zero-shot",
"BERT",
"PubMedBERT",
"en",
"dataset:ncbi_disease",
"dataset:bigbio/chemdner",
"dataset:bigbio/n2c2_2018_track2",
"dataset:bigbio/bc5cdr",
"dataset:bigbio/jnlpba",
"arxiv:2305.04928",
"lic... | 1,684 | 1,689 | 222 | 8 | ---
datasets:
- ncbi_disease
- bigbio/chemdner
- bigbio/n2c2_2018_track2
- bigbio/bc5cdr
- bigbio/jnlpba
language:
- en
library_name: transformers
license: mit
metrics:
- accuracy
- recall
- f1
- precision
tags:
- chemistry
- biology
- zero-shot
- BERT
- PubMedBERT
widget:
- text: Disease<SEP>Patient was diagnosed with... | [
"NAMED_ENTITY_RECOGNITION"
] | [
"BC5CDR",
"BIORED",
"CHEMDNER",
"JNLPBA",
"NCBI DISEASE"
] | BioNLP |
Snowflake/snowflake-arctic-embed-s | Snowflake | sentence-similarity | [
"sentence-transformers",
"onnx",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"arctic",
"snowflake-arctic-embed",
"transformers.js",
"arxiv:2407.18887",
"arxiv:2405.05374",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-embeddings-infe... | 1,712 | 1,733 | 29,658 | 19 | ---
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
- arctic
- snowflake-arctic-embed
- transformers.js
model-index:
- name: snowflake-snowflake-arctic-embed-s
results:
- task:
type: Classification
dataset:
name: M... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
himanshu23099/bge_embedding_finetune_v3 | himanshu23099 | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:3507",
"loss:GISTEmbedLoss",
"arxiv:1908.10084",
"arxiv:2402.16829",
"base_model:BAAI/bge-small-en-v1.5",
"base_model:finetune:BAAI/bge-small-en-v1.5",
"model-... | 1,731 | 1,732 | 12 | 0 | ---
base_model: BAAI/bge-small-en-v1.5
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@5
- cosine_ndcg@10
- cosine_ndcg@100
- cos... | [
"TEXT_CLASSIFICATION"
] | [
"CRAFT"
] | Non_BioNLP |
CAiRE/UniVaR-lambda-5 | CAiRE | sentence-similarity | [
"sentence-transformers",
"safetensors",
"nomic_bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"transformers",
"transformers.js",
"custom_code",
"en",
"arxiv:2402.01613",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_co... | 1,718 | 1,718 | 13 | 0 | ---
language:
- en
library_name: sentence-transformers
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- feature-extraction
- sentence-similarity
- mteb
- transformers
- transformers.js
model-index:
- name: epoch_0_model
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCou... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
RichardErkhov/EleutherAI_-_pythia-2.8b-deduped-8bits | RichardErkhov | text-generation | [
"transformers",
"safetensors",
"gpt_neox",
"text-generation",
"arxiv:2304.01373",
"arxiv:2101.00027",
"arxiv:2201.07311",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"8-bit",
"bitsandbytes",
"region:us"
] | 1,713 | 1,713 | 4 | 0 | ---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
pythia-2.8b-deduped - bnb 8bits
- Model creator: https://huggingface.co/EleutherAI/
- Original model: https://hug... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
aident-ai/bge-base-en-onnx | aident-ai | feature-extraction | [
"sentence-transformers",
"pytorch",
"onnx",
"bert",
"feature-extraction",
"sentence-similarity",
"transformers",
"mteb",
"en",
"arxiv:2309.07597",
"license:mit",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,694 | 1,695 | 19 | 2 | ---
language:
- en
license: mit
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
- mteb
---
This is a fork from [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) and exported to onnx for inference.
=======
<h1 align="center">FlagEmbedding</h1>
<h4 align=... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BEAR"
] | Non_BioNLP |
EleutherAI/pythia-1.4b | EleutherAI | text-generation | [
"transformers",
"pytorch",
"safetensors",
"gpt_neox",
"text-generation",
"causal-lm",
"pythia",
"en",
"dataset:EleutherAI/the_pile",
"arxiv:2304.01373",
"arxiv:2101.00027",
"arxiv:2201.07311",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compati... | 1,675 | 1,688 | 83,953 | 23 | ---
datasets:
- EleutherAI/the_pile
language:
- en
license: apache-2.0
tags:
- pytorch
- causal-lm
- pythia
---
The *Pythia Scaling Suite* is a collection of models developed to facilitate
interpretability research [(see paper)](https://arxiv.org/pdf/2304.01373.pdf).
It contains two sets of eight models of sizes
70... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
mxs980/gte-Qwen2-1.5B-instruct-Q8_0-GGUF | mxs980 | sentence-similarity | [
"sentence-transformers",
"gguf",
"mteb",
"transformers",
"Qwen2",
"sentence-similarity",
"llama-cpp",
"gguf-my-repo",
"base_model:Alibaba-NLP/gte-Qwen2-1.5B-instruct",
"base_model:quantized:Alibaba-NLP/gte-Qwen2-1.5B-instruct",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"... | 1,719 | 1,719 | 41 | 0 | ---
base_model: Alibaba-NLP/gte-Qwen2-1.5B-instruct
license: apache-2.0
tags:
- mteb
- sentence-transformers
- transformers
- Qwen2
- sentence-similarity
- llama-cpp
- gguf-my-repo
model-index:
- name: gte-qwen2-7B-instruct
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactu... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
medspaner/xlm-roberta-large-spanish-trials-misc-ents | medspaner | token-classification | [
"transformers",
"pytorch",
"xlm-roberta",
"token-classification",
"generated_from_trainer",
"license:cc-by-nc-4.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 1,705 | 1,727 | 12 | 0 | ---
license: cc-by-nc-4.0
metrics:
- precision
- recall
- f1
- accuracy
tags:
- generated_from_trainer
widget:
- text: Paciente normotenso (PA = 120/70 mmHg)
model-index:
- name: xlm-roberta-large-spanish-trials-cases-misc-ents
results: []
---
<!-- This model card has been generated automatically according to the in... | [
"NAMED_ENTITY_RECOGNITION"
] | [
"CT-EBM-SP",
"SCIELO"
] | BioNLP |
EleutherAI/pythia-160m-v0 | EleutherAI | text-generation | [
"transformers",
"pytorch",
"safetensors",
"gpt_neox",
"text-generation",
"causal-lm",
"pythia",
"pythia_v0",
"en",
"dataset:the_pile",
"arxiv:2101.00027",
"arxiv:2201.07311",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us... | 1,665 | 1,688 | 596 | 8 | ---
datasets:
- the_pile
language:
- en
license: apache-2.0
tags:
- pytorch
- causal-lm
- pythia
- pythia_v0
---
The *Pythia Scaling Suite* is a collection of models developed to facilitate
interpretability research. It contains two sets of eight models of sizes
70M, 160M, 410M, 1B, 1.4B, 2.8B, 6.9B, and 12B. For ea... | [
"QUESTION_ANSWERING",
"TRANSLATION"
] | [
"SCIQ"
] | Non_BioNLP |
LoneStriker/OpenBioLLM-Llama3-8B-6.0bpw-h6-exl2 | LoneStriker | text-generation | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-3",
"Mixtral",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"distillation",
"en",
"arxiv:2305.18290",
"arxiv:2303.13375",
"arxiv:2212.13138",
"arxiv:2305.09617",
"arxiv:2402.07023",
"base_model:meta-llama... | 1,714 | 1,714 | 7 | 0 | ---
base_model: meta-llama/Meta-Llama-3-8B
language:
- en
license: llama3
tags:
- llama-3
- llama
- Mixtral
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- distillation
widget:
- example_title: OpenBioLLM-8B
messages:
- role: system
content: You are an expert and experienced from the healthcare and biomedi... | [
"QUESTION_ANSWERING"
] | [
"MEDQA",
"PUBMEDQA"
] | BioNLP |
Nashhz/SBERT_KFOLD_JobDescriptions_Skills_UserPortfolios | Nashhz | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:16682",
"loss:CosineSimilarityLoss",
"arxiv:1908.10084",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,735 | 1,735 | 308 | 0 | ---
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:16682
- loss:CosineSimilarityLoss
widget:
- source_sentence: Hello, I am Redoan Ahmad I'm a professional Graphic Designer ... | [
"TEXT_CLASSIFICATION"
] | [
"CRAFT"
] | Non_BioNLP |
corto-ai/bge-reranker-large-onnx | corto-ai | feature-extraction | [
"transformers",
"onnx",
"xlm-roberta",
"text-classification",
"mteb",
"feature-extraction",
"en",
"zh",
"arxiv:2401.03462",
"arxiv:2312.15503",
"arxiv:2311.13534",
"arxiv:2310.07554",
"arxiv:2309.07597",
"license:mit",
"model-index",
"autotrain_compatible",
"text-embeddings-inference... | 1,717 | 1,717 | 11 | 0 | ---
language:
- en
- zh
license: mit
pipeline_tag: feature-extraction
tags:
- mteb
model-index:
- name: bge-reranker-base
results:
- task:
type: Reranking
dataset:
name: MTEB CMedQAv1
type: C-MTEB/CMedQAv1-reranking
config: default
split: test
revision: None
metrics:
... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BEAR"
] | Non_BioNLP |
croissantllm/CroissantLLMBase-GGUF | croissantllm | text-generation | [
"gguf",
"legal",
"code",
"text-generation-inference",
"art",
"text-generation",
"fr",
"en",
"dataset:cerebras/SlimPajama-627B",
"dataset:uonlp/CulturaX",
"dataset:pg19",
"dataset:bigcode/starcoderdata",
"dataset:croissantllm/croissant_dataset",
"arxiv:2402.00786",
"license:mit",
"endpo... | 1,707 | 1,714 | 64 | 4 | ---
datasets:
- cerebras/SlimPajama-627B
- uonlp/CulturaX
- pg19
- bigcode/starcoderdata
- croissantllm/croissant_dataset
language:
- fr
- en
license: mit
pipeline_tag: text-generation
tags:
- legal
- code
- text-generation-inference
- art
---
# CroissantLLM - Base GGUF (190k steps, Final version)
This model is part ... | [
"TRANSLATION"
] | [
"CRAFT"
] | Non_BioNLP |
Mihaiii/Squirtle | Mihaiii | sentence-similarity | [
"sentence-transformers",
"onnx",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"bge",
"mteb",
"dataset:Mihaiii/qa-assistant",
"license:mit",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,714 | 1,714 | 35 | 1 | ---
datasets:
- Mihaiii/qa-assistant
library_name: sentence-transformers
license: mit
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- bge
- mteb
model-index:
- name: Squirtle
results:
- task:
type: Classification
dataset:
name: MTEB Amazon... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | BioNLP |
Teradata/multilingual-e5-base | Teradata | sentence-similarity | [
"onnx",
"mteb",
"sentence-similarity",
"teradata",
"multilingual",
"af",
"am",
"ar",
"as",
"az",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"fi",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
... | 1,739 | 1,741 | 25 | 0 | ---
language:
- multilingual
- af
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- en
- eo
- es
- et
- eu
- fa
- fi
- fr
- fy
- ga
- gd
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- id
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
- lo
- lt
- lv
- mg
- mk
- ml
- mn
- mr
- ms
- my
-... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
t12e/instructor-base | t12e | sentence-similarity | [
"sentence-transformers",
"pytorch",
"t5",
"text-embedding",
"embeddings",
"information-retrieval",
"beir",
"text-classification",
"language-model",
"text-clustering",
"text-semantic-similarity",
"text-evaluation",
"prompt-retrieval",
"text-reranking",
"feature-extraction",
"sentence-si... | 1,685 | 1,685 | 9 | 0 | ---
language: en
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- text-embedding
- embeddings
- information-retrieval
- beir
- text-classification
- language-model
- text-clustering
- text-semantic-similarity
- text-evaluation
- prompt-retrieval
- text-reranking
- sentence-transformers
- feature-extraction... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
carsondial/slinger20241231-1 | carsondial | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:45000",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-ba... | 1,735 | 1,735 | 5 | 0 | ---
base_model: BAAI/bge-base-en-v1.5
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
... | [
"TEXT_CLASSIFICATION"
] | [
"BEAR"
] | Non_BioNLP |
aimlresearch2023/snowflake-arctic-embed-m-v1.5-Q8_0-GGUF | aimlresearch2023 | sentence-similarity | [
"sentence-transformers",
"gguf",
"feature-extraction",
"sentence-similarity",
"mteb",
"arctic",
"snowflake-arctic-embed",
"transformers.js",
"llama-cpp",
"gguf-my-repo",
"base_model:Snowflake/snowflake-arctic-embed-m-v1.5",
"base_model:quantized:Snowflake/snowflake-arctic-embed-m-v1.5",
"lic... | 1,726 | 1,726 | 7 | 0 | ---
base_model: Snowflake/snowflake-arctic-embed-m-v1.5
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
- arctic
- snowflake-arctic-embed
- transformers.js
- llama-cpp
- gguf-my-repo
model-index:
- name: snowflake-arctic-embed-m-v1.5
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
LoneStriker/OpenBioLLM-Llama3-70B-4.0bpw-h6-exl2 | LoneStriker | text-generation | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-3",
"Mixtral",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"distillation",
"conversational",
"en",
"arxiv:2305.18290",
"arxiv:2303.13375",
"arxiv:2212.13138",
"arxiv:2305.09617",
"arxiv:2402.07023",
"b... | 1,714 | 1,714 | 8 | 0 | ---
base_model: meta-llama/Meta-Llama-3-70B-Instruct
language:
- en
license: llama3
tags:
- llama-3
- llama
- Mixtral
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- distillation
widget:
- example_title: OpenBioLLM-70B
messages:
- role: system
content: You are an expert and experienced from the healthcare ... | [
"QUESTION_ANSWERING"
] | [
"MEDQA",
"PUBMEDQA"
] | BioNLP |
twadada/mpn-wordpca | twadada | null | [
"mteb",
"model-index",
"region:us"
] | 1,726 | 1,726 | 0 | 0 | ---
tags:
- mteb
model-index:
- name: mpnet_main_wordPCA
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: None
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
- type: accuracy
... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
gliner-community/gliner_xxl-v2.5 | gliner-community | token-classification | [
"gliner",
"pytorch",
"token-classification",
"multilingual",
"dataset:urchade/pile-mistral-v0.1",
"arxiv:2311.08526",
"license:apache-2.0",
"region:us"
] | 1,724 | 1,725 | 70 | 3 | ---
datasets:
- urchade/pile-mistral-v0.1
language:
- multilingual
library_name: gliner
license: apache-2.0
pipeline_tag: token-classification
---
# About
GLiNER is a Named Entity Recognition (NER) model capable of identifying any entity type using a bidirectional transformer encoder (BERT-like). It provides a practi... | [
"NAMED_ENTITY_RECOGNITION"
] | [
"ANATEM",
"BC5CDR"
] | Non_BioNLP |
twadada/nmc-300-w50k-b30k | twadada | null | [
"mteb",
"model-index",
"region:us"
] | 1,726 | 1,726 | 0 | 0 | ---
tags:
- mteb
model-index:
- name: nomic_classification_307_w50k_b30k
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: None
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
- t... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
PlanTL-GOB-ES/bsc-bio-ehr-es | PlanTL-GOB-ES | fill-mask | [
"transformers",
"pytorch",
"roberta",
"fill-mask",
"biomedical",
"clinical",
"ehr",
"spanish",
"es",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 1,649 | 1,668 | 665 | 12 | ---
language:
- es
license: apache-2.0
metrics:
- ppl
tags:
- biomedical
- clinical
- ehr
- spanish
widget:
- text: El único antecedente personal a reseñar era la <mask> arterial.
- text: Las radiologías óseas de cuerpo entero no detectan alteraciones <mask>, ni
alteraciones vertebrales.
- text: En el <mask> toraco... | [
"NAMED_ENTITY_RECOGNITION",
"TEXT_CLASSIFICATION"
] | [
"CANTEMIST",
"PHARMACONER",
"SCIELO"
] | BioNLP |
barisaydin/bge-small-en | barisaydin | feature-extraction | [
"transformers",
"pytorch",
"safetensors",
"bert",
"feature-extraction",
"mteb",
"sentence transformers",
"en",
"arxiv:2309.07597",
"license:mit",
"model-index",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 1,695 | 1,695 | 9 | 0 | ---
language:
- en
license: mit
tags:
- mteb
- sentence transformers
model-index:
- name: bge-small-en
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: mteb/amazon_counterfactual
config: en
split: test
revision: e8379541a... | [
"SEMANTIC_SIMILARITY",
"SUMMARIZATION"
] | [
"BEAR",
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
bnightning/gte-Qwen2-7B-instruct-Q4_K_M-GGUF | bnightning | sentence-similarity | [
"sentence-transformers",
"gguf",
"mteb",
"transformers",
"Qwen2",
"sentence-similarity",
"llama-cpp",
"gguf-my-repo",
"base_model:Alibaba-NLP/gte-Qwen2-7B-instruct",
"base_model:quantized:Alibaba-NLP/gte-Qwen2-7B-instruct",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endp... | 1,741 | 1,741 | 12 | 0 | ---
base_model: Alibaba-NLP/gte-Qwen2-7B-instruct
license: apache-2.0
tags:
- mteb
- sentence-transformers
- transformers
- Qwen2
- sentence-similarity
- llama-cpp
- gguf-my-repo
model-index:
- name: gte-qwen2-7B-instruct
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactual... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
twadada/nmc-nignore15 | twadada | null | [
"mteb",
"model-index",
"region:us"
] | 1,726 | 1,726 | 0 | 0 | ---
tags:
- mteb
model-index:
- name: nomic_classification_nignore15
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassification (en)
type: None
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
- type:... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
tensorblock/gte-Qwen2-7B-instruct-GGUF | tensorblock | sentence-similarity | [
"sentence-transformers",
"gguf",
"mteb",
"transformers",
"Qwen2",
"sentence-similarity",
"TensorBlock",
"GGUF",
"base_model:Alibaba-NLP/gte-Qwen2-7B-instruct",
"base_model:quantized:Alibaba-NLP/gte-Qwen2-7B-instruct",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_... | 1,731 | 1,731 | 3,236 | 8 | ---
base_model: Alibaba-NLP/gte-Qwen2-7B-instruct
license: apache-2.0
tags:
- mteb
- sentence-transformers
- transformers
- Qwen2
- sentence-similarity
- TensorBlock
- GGUF
model-index:
- name: gte-qwen2-7B-instruct
results:
- task:
type: Classification
dataset:
name: MTEB AmazonCounterfactualClassi... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
ymelka/camembert-cosmetic-similarity-v2 | ymelka | sentence-similarity | [
"sentence-transformers",
"safetensors",
"camembert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:5000",
"loss:CoSENTLoss",
"arxiv:1908.10084",
"base_model:ymelka/camembert-cosmetic-finetuned",
"base_model:finetune:ymelka/camembert-cosmetic-finetuned",
"... | 1,718 | 1,718 | 10 | 0 | ---
base_model: ymelka/camembert-cosmetic-finetuned
datasets: []
language: []
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
pipeline_tag: sentence... | [
"TEXT_CLASSIFICATION",
"SEMANTIC_SIMILARITY"
] | [
"CAS"
] | Non_BioNLP |
CyberOptic/multilingual-e5-large-Q8_0-GGUF | CyberOptic | feature-extraction | [
"sentence-transformers",
"gguf",
"mteb",
"Sentence Transformers",
"sentence-similarity",
"feature-extraction",
"llama-cpp",
"gguf-my-repo",
"multilingual",
"af",
"am",
"ar",
"as",
"az",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"cs",
"cy",
"da",
"de",
"el",
"en",
"eo",
... | 1,741 | 1,741 | 8 | 0 | ---
base_model: intfloat/multilingual-e5-large
language:
- multilingual
- af
- am
- ar
- as
- az
- be
- bg
- bn
- br
- bs
- ca
- cs
- cy
- da
- de
- el
- en
- eo
- es
- et
- eu
- fa
- fi
- fr
- fy
- ga
- gd
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- id
- is
- it
- ja
- jv
- ka
- kk
- km
- kn
- ko
- ku
- ky
- la
- lo
- l... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
LoneStriker/OpenBioLLM-Llama3-8B-5.0bpw-h6-exl2 | LoneStriker | text-generation | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-3",
"Mixtral",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"distillation",
"en",
"arxiv:2305.18290",
"arxiv:2303.13375",
"arxiv:2212.13138",
"arxiv:2305.09617",
"arxiv:2402.07023",
"base_model:meta-llama... | 1,714 | 1,714 | 8 | 0 | ---
base_model: meta-llama/Meta-Llama-3-8B
language:
- en
license: llama3
tags:
- llama-3
- llama
- Mixtral
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- distillation
widget:
- example_title: OpenBioLLM-8B
messages:
- role: system
content: You are an expert and experienced from the healthcare and biomedi... | [
"QUESTION_ANSWERING"
] | [
"MEDQA",
"PUBMEDQA"
] | BioNLP |
hkunlp/instructor-base | hkunlp | sentence-similarity | [
"sentence-transformers",
"pytorch",
"t5",
"text-embedding",
"embeddings",
"information-retrieval",
"beir",
"text-classification",
"language-model",
"text-clustering",
"text-semantic-similarity",
"text-evaluation",
"prompt-retrieval",
"text-reranking",
"feature-extraction",
"sentence-si... | 1,671 | 1,674 | 11,563 | 116 | ---
language: en
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- text-embedding
- embeddings
- information-retrieval
- beir
- text-classification
- language-model
- text-clustering
- text-semantic-similarity
- text-evaluation
- prompt-retrieval
- text-reranking
- sentence-transformers
- feature-extraction... | [
"SUMMARIZATION"
] | [
"BIOSSES",
"SCIFACT"
] | Non_BioNLP |
BSC-LT/salamandra-7b-instruct | BSC-LT | text-generation | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"bg",
"ca",
"code",
"cs",
"cy",
"da",
"de",
"el",
"en",
"es",
"et",
"eu",
"fi",
"fr",
"ga",
"gl",
"hr",
"hu",
"it",
"lt",
"lv",
"mt",
"nl",
"nn",
"oc",
"pl",
"pt",
"ro",
"ru... | 1,727 | 1,740 | 12,988 | 54 | ---
base_model:
- BSC-LT/salamandra-7b
datasets:
- oscar-corpus/colossal-oscar-1.0
- HuggingFaceFW/fineweb-edu
- joelniklaus/eurlex_resources
- joelito/legal-mc4
- projecte-aina/CATalog
- UFRGS/brwac
- community-datasets/hrwac
- danish-foundation-models/danish-gigaword
- HiTZ/euscrawl
- PleIAs/French-PD-Newspapers
- Pl... | [
"QUESTION_ANSWERING",
"TRANSLATION",
"SUMMARIZATION",
"PARAPHRASING"
] | [
"BEAR",
"SCIELO"
] | Non_BioNLP |
BSC-NLP4BIA/bsc-bio-ehr-es-meddocan | BSC-NLP4BIA | token-classification | [
"transformers",
"pytorch",
"roberta",
"token-classification",
"biomedical",
"clinical",
"EHR",
"spanish",
"anonymization",
"es",
"base_model:PlanTL-GOB-ES/bsc-bio-ehr-es",
"base_model:finetune:PlanTL-GOB-ES/bsc-bio-ehr-es",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"... | 1,717 | 1,722 | 15 | 0 | ---
base_model:
- PlanTL-GOB-ES/bsc-bio-ehr-es
language:
- es
license: apache-2.0
metrics:
- precision
- recall
- f1
tags:
- biomedical
- clinical
- EHR
- spanish
- anonymization
widget:
- text: El diagnóstico definitivo de nuestro paciente fue de un Adenocarcinoma de
pulmón cT2a cN3 cM1a Estadio IV (por una única ... | [
"NAMED_ENTITY_RECOGNITION"
] | [
"MEDDOCAN"
] | BioNLP |
samchain/econo-sentence-v2 | samchain | sentence-similarity | [
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:45044",
"loss:CoSENTLoss",
"economics",
"finance",
"en",
"dataset:samchain/econo-pairs-v2",
"arxiv:1908.10084",
"base_model:samchain/EconoBert",
"base_mode... | 1,739 | 1,739 | 64 | 2 | ---
base_model: samchain/EconoBert
datasets:
- samchain/econo-pairs-v2
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- pearson_cosine
- spearman_cosine
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- ... | [
"TEXT_CLASSIFICATION",
"SEMANTIC_SIMILARITY"
] | [
"BEAR"
] | Non_BioNLP |
AventIQ-AI/bert-medical-entity-extraction | AventIQ-AI | null | [
"safetensors",
"bert",
"region:us"
] | 1,740 | 1,740 | 72 | 3 | ---
{}
---
# Medical Entity Extraction with BERT
## 📌 Overview
This repository hosts the quantized version of the `bert-base-cased` model for Medical Entity Extraction using the 'tner/bc5cdr' dataset. The model is specifically designed to recognize entities related to **Disease,Symptoms,Drug**. The model has been opt... | [
"NAMED_ENTITY_RECOGNITION"
] | [
"BC5CDR"
] | BioNLP |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.