{"id":40,"date":"2024-12-02T11:16:09","date_gmt":"2024-12-02T09:16:09","guid":{"rendered":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/?p=40"},"modified":"2024-12-02T11:16:10","modified_gmt":"2024-12-02T09:16:10","slug":"what-are-embeddings-produced-by-llms","status":"publish","type":"post","link":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/","title":{"rendered":"What are embeddings produced by LLMs?"},"content":{"rendered":"\n<p>An ordinary user interacts with large language models, like ChatGPT, by writing prompts through the user interface. In addition to this, large language models offer another functionality for technically skilled users \u2013 the creation of <a href=\"https:\/\/www.ibm.com\/topics\/embedding\" target=\"_blank\" rel=\"noreferrer noopener\">embeddings<\/a> based on text. But what exactly are these embeddings, and what are they used for?<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">Meaning of text in vectors<\/h1>\n\n\n\n<p>When a large language model is given some text to embed, it produces a vector as a result. A vector is a list of numbers that may not be immediately interpretable to the human eye, but it enables the exploration of the text&#8217;s meaning through mathematical methods. <strong>These vectors produced by the language model are called embeddings.<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/github.com\/mikahama\/uralicNLP\/\">UralicNLP Python library<\/a> provides tools for embedding text using different language models. Here is an example of how text can be embedded with OpenAI&#8217;s model using UralicNLP.<\/p>\n\n\n\n<p style=\"margin-left:64px\">from uralicNLP.llm import get_llm<br>llm = get_llm(&#8221;chatgpt&#8221;, &#8221;VAIHDA T\u00c4H\u00c4N API-AVAIMESI&#8221;, model=&#8221;text-embedding-3-small&#8221;)<br>llm.embed(&#8221;Teksti, jonka haluat upottaa&#8221;)<br>&gt;&gt;[-0.1803697, 1.1973963, 0.5283669, 1.5049516, -0.27077377&#8230;]<\/p>\n\n\n\n<p>As seen in the example, the result of an embedding is a list of numbers. These numbers represent the meaning of the text and can be used to compare the similarity of texts through mathematical methods.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">What are the benefits of embeddings?<\/h1>\n\n\n\n<p>With embeddings, large volumes of text can be stored in a vector database for quick retrieval. This means database searches are based on meaning rather than character strings. The most common use case for such vector databases currently is the <a href=\"https:\/\/blogs.nvidia.com\/blog\/what-is-retrieval-augmented-generation\/\" target=\"_blank\" rel=\"noreferrer noopener\">RAG model<\/a>.<\/p>\n\n\n\n<p>RAG stands for Retrieval-Augmented Generation, which refers to a process where a large language model is provided with not just the user prompt but also source material to help generate a response. Retrieving the source material involves using embeddings to find documents relevant to the user\u2019s input from a vector database. For example, <a href=\"https:\/\/mikko.metropolia.fi\/\">Metropolia\u2019s own Mikro-Mikko<\/a> operates based on this principle.<\/p>\n\n\n\n<p>Embeddings can also be used to automatically group text documents into clusters of similar texts. This can be done with UralicNLP as follows.<\/p>\n\n\n\n<p style=\"margin-left: 64px\">from uralicNLP.llm import get_llm<br>from uralicNLP import semantics<br>llm = get_llm(&#8221;chatgpt&#8221;, &#8221;VAIHDA T\u00c4H\u00c4N API-AVAIMESI&#8221;, model=&#8221;text-embedding-3-small&#8221;)<br>texts = [\u201ckoirat on hauskoja\u201d, \u201cautot ajaa nopeasti\u201d, \u201ckissat leikkii kesken\u00e4\u00e4n\u201d, \u201crekat ajaa kaupungista toiseen\u201d]<br>semantics.cluster(texts, llm)<br>&gt;&gt;[[\u201ckoirat on hauskoja\u201d, \u201ckissat leikkii kesken\u00e4\u00e4n\u201d], [\u201cautot ajaa nopeasti\u201d, \u201crekat ajaa kaupungista toiseen\u201d]]<\/p>\n\n\n\n<p>The result is that texts are grouped into clusters of similar texts using embeddings and calculating their similarity.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">Does the model matter when embedding?<\/h1>\n\n\n\n<p>Embeddings can be generated using both commercial large language models and open-source language models. When choosing a model, it\u2019s important to remember that embeddings are not compatible across models. For example, you cannot create some embeddings with OpenAI&#8217;s GPT-4 and others with an open-source LLaMA model and expect them to work together. Each model has learned its own representation of meaning from its training data, so the numerical content of the embeddings varies between models.<\/p>\n\n\n\n<p>When choosing a model, it&#8217;s important to consider the cost of the model, the languages it supports, and its context window. Larger models can accommodate a large amount of text within the context window, allowing for a single embedding of an entire text. Smaller models require the text to be split into segments. This technical limitation can be significant depending on how the embeddings are intended to be used.<\/p>\n\n\n\n<p>Not all models support all languages. If a language model produces poor Finnish responses to prompts, it likely does not understand Finnish very well. Consequently, embeddings generated for Finnish text may not capture the meaning accurately enough.<\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>An ordinary user interacts with large language models, like ChatGPT, by writing prompts through the user interface. In addition to this, large language models offer another functionality for technically skilled users \u2013 the creation of embeddings based on text. But what exactly are these embeddings, and what are they used for? Meaning of text in [&hellip;]<\/p>\n","protected":false},"author":93,"featured_media":39,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_exactmetrics_skip_tracking":false,"_exactmetrics_sitenote_active":false,"_exactmetrics_sitenote_note":"","_exactmetrics_sitenote_category":0,"footnotes":""},"categories":[22],"tags":[],"class_list":["post-40","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-yleinen-en"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.3 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>What are embeddings produced by LLMs? - Teko\u00e4ly\u00e4 tutkimassa<\/title>\n<meta name=\"description\" content=\"Text can turned into an embedding using large language models. What does this mean in practice, and what are the benefits of it?\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/\" \/>\n<meta property=\"og:locale\" content=\"fi_FI\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"What are embeddings produced by LLMs? - Teko\u00e4ly\u00e4 tutkimassa\" \/>\n<meta property=\"og:description\" content=\"Text can turned into an embedding using large language models. What does this mean in practice, and what are the benefits of it?\" \/>\n<meta property=\"og:url\" content=\"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/\" \/>\n<meta property=\"og:site_name\" content=\"Teko\u00e4ly\u00e4 tutkimassa\" \/>\n<meta property=\"article:published_time\" content=\"2024-12-02T09:16:09+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-12-02T09:16:10+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-content\/uploads\/sites\/29\/2024\/12\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"2560\" \/>\n\t<meta property=\"og:image:height\" content=\"1463\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Mika H\u00e4m\u00e4l\u00e4inen\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Kirjoittanut\" \/>\n\t<meta name=\"twitter:data1\" content=\"Mika H\u00e4m\u00e4l\u00e4inen\" \/>\n\t<meta name=\"twitter:label2\" content=\"Arvioitu lukuaika\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minuuttia\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/\"},\"author\":{\"name\":\"Mika H\u00e4m\u00e4l\u00e4inen\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/#\\\/schema\\\/person\\\/3a4e7defe237b70cb74b463446adacbf\"},\"headline\":\"What are embeddings produced by LLMs?\",\"datePublished\":\"2024-12-02T09:16:09+00:00\",\"dateModified\":\"2024-12-02T09:16:10+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/\"},\"wordCount\":607,\"commentCount\":1,\"image\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/wp-content\\\/uploads\\\/sites\\\/29\\\/2024\\\/12\\\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg\",\"inLanguage\":\"fi\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/\",\"url\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/\",\"name\":\"What are embeddings produced by LLMs? - Teko\u00e4ly\u00e4 tutkimassa\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/wp-content\\\/uploads\\\/sites\\\/29\\\/2024\\\/12\\\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg\",\"datePublished\":\"2024-12-02T09:16:09+00:00\",\"dateModified\":\"2024-12-02T09:16:10+00:00\",\"author\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/#\\\/schema\\\/person\\\/3a4e7defe237b70cb74b463446adacbf\"},\"description\":\"Text can turned into an embedding using large language models. What does this mean in practice, and what are the benefits of it?\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#breadcrumb\"},\"inLanguage\":\"fi\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fi\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#primaryimage\",\"url\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/wp-content\\\/uploads\\\/sites\\\/29\\\/2024\\\/12\\\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg\",\"contentUrl\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/wp-content\\\/uploads\\\/sites\\\/29\\\/2024\\\/12\\\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg\",\"width\":2560,\"height\":1463,\"caption\":\"Taikurin hattu, jonne putoaa A-kirjain ja joukko numeroita\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/en\\\/2024\\\/12\\\/02\\\/what-are-embeddings-produced-by-llms\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Etusivu\",\"item\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"What are embeddings produced by LLMs?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/#website\",\"url\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/\",\"name\":\"Teko\u00e4ly\u00e4 tutkimassa\",\"description\":\"Tieteeseen perustuvaa tietoa teko\u00e4lyst\u00e4\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fi\"},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/#\\\/schema\\\/person\\\/3a4e7defe237b70cb74b463446adacbf\",\"name\":\"Mika H\u00e4m\u00e4l\u00e4inen\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fi\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ef863f72fd4d3cae89a7b4f76167d31e2ff8b2869d30d1fbd4ede7a2ab86a089?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ef863f72fd4d3cae89a7b4f76167d31e2ff8b2869d30d1fbd4ede7a2ab86a089?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ef863f72fd4d3cae89a7b4f76167d31e2ff8b2869d30d1fbd4ede7a2ab86a089?s=96&d=mm&r=g\",\"caption\":\"Mika H\u00e4m\u00e4l\u00e4inen\"},\"description\":\"on kieliteknologian tohtori ja h\u00e4n toimii AI-projektip\u00e4\u00e4llikk\u00f6n\u00e4 Metropoliassa. H\u00e4nen viimeaikaiset tutkimusintressins\u00e4 ovat kieliteknologian kehitt\u00e4minen uhanalaisille kielille sek\u00e4 kieliteknologian ja digihumanismin yhteensovittaminen.\",\"sameAs\":[\"https:\\\/\\\/peoplefinder.metropolia.fi\\\/fi\\\/profile\\\/325063\\\/staff\\\/Mika-Hamalainen\"],\"url\":\"https:\\\/\\\/blogit.metropolia.fi\\\/tekoalya-tutkimassa\\\/author\\\/mikakha\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"What are embeddings produced by LLMs? - Teko\u00e4ly\u00e4 tutkimassa","description":"Text can turned into an embedding using large language models. What does this mean in practice, and what are the benefits of it?","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/","og_locale":"fi_FI","og_type":"article","og_title":"What are embeddings produced by LLMs? - Teko\u00e4ly\u00e4 tutkimassa","og_description":"Text can turned into an embedding using large language models. What does this mean in practice, and what are the benefits of it?","og_url":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/","og_site_name":"Teko\u00e4ly\u00e4 tutkimassa","article_published_time":"2024-12-02T09:16:09+00:00","article_modified_time":"2024-12-02T09:16:10+00:00","og_image":[{"width":2560,"height":1463,"url":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-content\/uploads\/sites\/29\/2024\/12\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg","type":"image\/jpeg"}],"author":"Mika H\u00e4m\u00e4l\u00e4inen","twitter_card":"summary_large_image","twitter_misc":{"Kirjoittanut":"Mika H\u00e4m\u00e4l\u00e4inen","Arvioitu lukuaika":"4 minuuttia"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#article","isPartOf":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/"},"author":{"name":"Mika H\u00e4m\u00e4l\u00e4inen","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/#\/schema\/person\/3a4e7defe237b70cb74b463446adacbf"},"headline":"What are embeddings produced by LLMs?","datePublished":"2024-12-02T09:16:09+00:00","dateModified":"2024-12-02T09:16:10+00:00","mainEntityOfPage":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/"},"wordCount":607,"commentCount":1,"image":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#primaryimage"},"thumbnailUrl":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-content\/uploads\/sites\/29\/2024\/12\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg","inLanguage":"fi","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/","url":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/","name":"What are embeddings produced by LLMs? - Teko\u00e4ly\u00e4 tutkimassa","isPartOf":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/#website"},"primaryImageOfPage":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#primaryimage"},"image":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#primaryimage"},"thumbnailUrl":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-content\/uploads\/sites\/29\/2024\/12\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg","datePublished":"2024-12-02T09:16:09+00:00","dateModified":"2024-12-02T09:16:10+00:00","author":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/#\/schema\/person\/3a4e7defe237b70cb74b463446adacbf"},"description":"Text can turned into an embedding using large language models. What does this mean in practice, and what are the benefits of it?","breadcrumb":{"@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#breadcrumb"},"inLanguage":"fi","potentialAction":[{"@type":"ReadAction","target":["https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/"]}]},{"@type":"ImageObject","inLanguage":"fi","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#primaryimage","url":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-content\/uploads\/sites\/29\/2024\/12\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg","contentUrl":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-content\/uploads\/sites\/29\/2024\/12\/Firefly-A-B-C-gets-converted-into-1-2-3-in-a-magical-way-through-a-magicians-hat-25811-scaled.jpg","width":2560,"height":1463,"caption":"Taikurin hattu, jonne putoaa A-kirjain ja joukko numeroita"},{"@type":"BreadcrumbList","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/en\/2024\/12\/02\/what-are-embeddings-produced-by-llms\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Etusivu","item":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/"},{"@type":"ListItem","position":2,"name":"What are embeddings produced by LLMs?"}]},{"@type":"WebSite","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/#website","url":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/","name":"Teko\u00e4ly\u00e4 tutkimassa","description":"Tieteeseen perustuvaa tietoa teko\u00e4lyst\u00e4","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fi"},{"@type":"Person","@id":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/#\/schema\/person\/3a4e7defe237b70cb74b463446adacbf","name":"Mika H\u00e4m\u00e4l\u00e4inen","image":{"@type":"ImageObject","inLanguage":"fi","@id":"https:\/\/secure.gravatar.com\/avatar\/ef863f72fd4d3cae89a7b4f76167d31e2ff8b2869d30d1fbd4ede7a2ab86a089?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/ef863f72fd4d3cae89a7b4f76167d31e2ff8b2869d30d1fbd4ede7a2ab86a089?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/ef863f72fd4d3cae89a7b4f76167d31e2ff8b2869d30d1fbd4ede7a2ab86a089?s=96&d=mm&r=g","caption":"Mika H\u00e4m\u00e4l\u00e4inen"},"description":"on kieliteknologian tohtori ja h\u00e4n toimii AI-projektip\u00e4\u00e4llikk\u00f6n\u00e4 Metropoliassa. H\u00e4nen viimeaikaiset tutkimusintressins\u00e4 ovat kieliteknologian kehitt\u00e4minen uhanalaisille kielille sek\u00e4 kieliteknologian ja digihumanismin yhteensovittaminen.","sameAs":["https:\/\/peoplefinder.metropolia.fi\/fi\/profile\/325063\/staff\/Mika-Hamalainen"],"url":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/author\/mikakha\/"}]}},"lang":"en","translations":{"en":40,"fi":33},"pll_sync_post":[],"_links":{"self":[{"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/posts\/40","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/users\/93"}],"replies":[{"embeddable":true,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/comments?post=40"}],"version-history":[{"count":5,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/posts\/40\/revisions"}],"predecessor-version":[{"id":45,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/posts\/40\/revisions\/45"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/media\/39"}],"wp:attachment":[{"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/media?parent=40"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/categories?post=40"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogit.metropolia.fi\/tekoalya-tutkimassa\/wp-json\/wp\/v2\/tags?post=40"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}