(docs) Semrush audit issues (#65)

* fix page has 1 links to redirect chains or loops

* fix duplicate meta descriptions

fern/pages/llm-university/intro-semantic-search/further-reading-search.mdx

@@ -8,6 +8,6 @@ updatedAt: "Wed Apr 03 2024 18:49:18 GMT+0000 (Coordinated Universal Time)"
 ---
 We hope you enjoyed learning about the different search methods you can use with large language models! If you'd like to delve more into search, please check out the following posts.
 
-- [Text embeddings visually explained](https://txt.cohere.com/text-embeddings/): A visual approach to text embeddings, including applications in search, clustering, and classification.
+- [Text embeddings visually explained](https://cohere.com/llmu/text-embeddings/): A visual approach to text embeddings, including applications in search, clustering, and classification.
 - [Semantic Search](/docs/semantic-search): Build a simple semantic search engine with a dataset of questions.
 - [Multilingual Movie Search](/page/multilingual-movie-search): Build your own movie search app. As input, you describe what movie you'd like to watch, and the model suggest a list of movies that match that description, in multiple languages.

fern/pages/llm-university/intro-text-representation/classification-using-embeddings.mdx

@@ -70,13 +70,13 @@ print(f"Prediction accuracy is {100*score}%")
 Here we can see that all predictions match the actual classes:
 
 
-<img src='../../../assets/images/bc5cfb4-image.png' alt='color)' />
+<img src='../../../assets/images/bc5cfb4-image.png' alt='two graphs where all predictions match the actual classes' />
 
 ### Conclusion
 
 As you can see, there are different ways to use the endpoints to do classification! As you learned before, you can use the Classify endpoint, but you can also use the Embed endpoint and train a simple classifier on the resulting data. Since embeddings capture context so well, the job of the classifier is vastly simplified, and this is why we could train a simple SVM.
 
 ### Original Source
 
-This material comes from the post <a target="_blank" href="https://txt.cohere.com/text-embeddings/">Text Embeddings Visually Explained</a>
+This material comes from the post <a target="_blank" href="https://cohere.com/llmu/text-embeddings/">Text Embeddings Visually Explained</a>
 

fern/pages/llm-university/intro-text-representation/clustering-using-embeddings.mdx

@@ -54,4 +54,4 @@ In this chapter, you learned how to cluster a dataset of sentences, and you obse
 
 ### Original Source
 
-This material comes from the post <a target="_blank" href="https://txt.cohere.com/text-embeddings/">Text Embeddings Visually Explained</a>
+This material comes from the post <a target="_blank" href="https://cohere.com/llmu/text-embeddings/">Text Embeddings Visually Explained</a>

fern/pages/llm-university/intro-text-representation/embeddings-visualizing-data.mdx

@@ -89,4 +89,4 @@ So now that we’ve seen how embeddings can represent text, in the following cha
 
 ### Original Source
 
-This material comes from the post <a target="_blank" href="https://txt.cohere.com/text-embeddings/">Text Embeddings Visually Explained</a>
+This material comes from the post <a target="_blank" href="https://cohere.com/llmu/text-embeddings/">Text Embeddings Visually Explained</a>

fern/pages/llm-university/intro-text-representation/semantic-search-using-embeddings.mdx

@@ -69,4 +69,4 @@ There are many more applications of embeddings, which you'll learn in the follow
 
 ### Original Source
 
-This material comes from the post <a target="_blank" href="https://txt.cohere.com/text-embeddings/">Text Embeddings Visually Explained</a>
+This material comes from the post <a target="_blank" href="https://cohere.com/llmu/text-embeddings/">Text Embeddings Visually Explained</a>

fern/pages/llm-university/sandbox/chapter-2-text-embeddings.mdx

@@ -1,7 +1,7 @@
 ---
 title: "REMOVE Text Embeddings Visually Explained"
 slug: "docs/chapter-2-text-embeddings"
-subtitle: "Post:https://txt.cohere.com/text-embeddings/ And lab https://github.com/cohere-ai/notebooks/blob/main/notebooks/Visualizing_Text_Embeddings.ipynb"
+subtitle: "Post:https://cohere.com/llmu/text-embeddings/ And lab https://github.com/cohere-ai/notebooks/blob/main/notebooks/Visualizing_Text_Embeddings.ipynb"
 hidden: true
 createdAt: "Tue Apr 25 2023 23:32:39 GMT+0000 (Coordinated Universal Time)"
 updatedAt: "Mon Oct 23 2023 14:40:59 GMT+0000 (Coordinated Universal Time)"

fern/pages/llm-university/sandbox/chapter-5-building-a-classifier-with-the-cohere-api.mdx

@@ -149,7 +149,7 @@ In this section, we’ll look at option number 2, which is how we can utilize th
 
 This time, we’ll use the full 800 training data points that we prepared earlier. The first step is to turn the training and test inputs (i.e., the airline queries) into embeddings.
 
-What exactly are embeddings? You can read [more about it in this chapter](https://txt.cohere.com/text-embeddings/), but in summary, it is a set of numbers that represent what a piece of text means, capturing its context and semantics. In our case, we are using the medium model, which generates embeddings of size 2096. So for each text we input, we will get an embedding consisting of 2096 numbers, which looks like the following:
+What exactly are embeddings? You can read [more about it in this chapter](https://cohere.com/llmu/text-embeddings/), but in summary, it is a set of numbers that represent what a piece of text means, capturing its context and semantics. In our case, we are using the medium model, which generates embeddings of size 2096. So for each text we input, we will get an embedding consisting of 2096 numbers, which looks like the following:
 
 **[0.20641953, 0.35582256, 0.6058123, -0.058944624, 0.8949609, 1.2956009, 1.2408538, -0.89241934, -0.56218493, -0.5521631, -0.11521566, 0.9081634, 1.662983, -0.9293592, -2.3170912, 1.177852, 0.35577637, ... ]**
 

fern/pages/text-embeddings/embeddings.mdx

@@ -3,7 +3,7 @@ title: "Introduction to Embeddings at Cohere"
 slug: "docs/embeddings"
 
 hidden: false
-description: "Use the API to generate completions, distill text into semantically meaningful vectors, and more. Get state-of-the-art natural language processing without the need for expensive supercomputing infrastructure."
+description: "Embeddings convert text into numerical representations to measure similarity, useful for tasks like search and classification, with support for multiple languages and various compression types."
 image: "../../assets/images/fa074c3-cohere_docs_preview_image_1200x630_copy.jpg"  
 keywords: "vector embeddings, embeddings, natural language processing"
 

fern/pages/text-generation/documents-and-citations.mdx

@@ -101,4 +101,4 @@ Also, we can clearly see that it _has_ utilized the document. Our first document
 
 Finally, note that the output contains a citations object that tells us not only which documents the model relied upon (with the `"text"` and `“document_ids"` fields), but also the particular part of the claim supported by a particular document (with the `“start”` and `“end”` fields, which are spans that tell us the location of the supported claim inside the reply). This citation object is included because the model was able to use the documents provided, but if it hadn’t been able to do so, no citation object would be present.
 
-You can experiment with RAG in the [chat playground](https://dashboard.cohere.com/playground/?_gl=1*1jqpo29*_ga*MTAxNTg1NTM1MS4xNjk1MjMwODQw*_ga_CRGS116RZS*MTY5NzIxNDAxNC40My4xLjE2OTcyMTQwMTQuNjAuMC4w).
+You can experiment with RAG in the [chat playground](https://dashboard.cohere.com/playground/chat).