Tech GPT

vector partitioning in Pinecone using multiple indexes, along with an example use case. 🌟 Multi-Tenancy and Efficient Querying with Namespaces What Is Multi-Tenancy? Multi-tenancy is a software architecture pattern where a single system serves multiple customers (tenants) simultaneously. Each tenant’s data is isolated to ensure privacy and security. Pinecone’s abstractions (indexes, namespaces, and metadata) make building multi-tenant systems straightforward. Namespaces for Data Isolation: Pinecone allows you to partition vectors into namespaces within an index. Each namespace contains related vectors for a specific tenant. Queries and other operations are limited to one namespace at a time. Data isolation enhances query performance by separating data segments. Namespaces scale independently, ensuring efficient operations even for different workloads. Example Use Case: SmartWiki’s AI-Assisted Wiki: Scenario: SmartWiki serves millions of companies and individuals. Each customer (tenant...

Let’s explore the concept of namespaces in Pinecone’s vector database! 🌟🔍 Namespaces in Pinecone: Organizing Vectors with Style 📁 What Are Namespaces? Namespaces allow you to partition the vectors in an index. Each namespace acts like a separate container for related vectors. Queries and other operations are then limited to one specific namespace. Think of it as organizing your vector data into different labeled folders. Why Use Namespaces? Optimized Search: By dividing your vectors into namespaces, you can focus searches on specific subsets. For example, you might want one namespace for articles by content and another for articles by title. Contextual Filtering: Metadata or context-specific vectors can reside in different namespaces. This helps you filter and retrieve relevant information efficiently. Example Use Case : Coffee Shop Locator Bot ☕🤖: Imagine you’re building a chatbot that finds nearby coffee shops. You have two namespaces: Namespace 1 (“ns1”): Contains vectors for c...

What Is Metadata? Metadata refers to additional information associated with each vector in the database. It provides context, labels, or attributes for the vectors. Think of it as “extra data” that helps you organize and filter your vectors effectively. Difference Between Vector Indexing and Metadata: Vector Indexing: Vector indexing focuses on the vectors themselves. It allows you to perform similarity searches, retrieve vectors, and manage CRUD (Create, Read, Update, Delete) operations. The primary goal is efficient retrieval based on vector similarity. Metadata: Metadata complements vector indexing. It adds descriptive information to each vector. You can filter vectors based on metadata attributes. Metadata enables more specific queries and context-aware searches. Use Cases and Examples: Movie Recommendations: Imagine you’re building a movie recommendation system. Each movie vector has metadata like genre (e.g., “comedy,” “action,” “documentary”). When a user searches for “comedy mo...

Pinecone’s serverless indexing, its use cases! 🌟🚀 Pinecone Serverless Indexing Pinecone’s serverless indexing is a powerful feature that allows you to create and manage indexes without worrying about infrastructure setup or scaling. Here’s what you need to know: What Is It? A serverless index automatically scales based on usage. You pay only for the data stored and operations performed. No need to configure compute or storage resources. Ideal for organizations on the Standard and Enterprise plans. Use Cases: Semantic Search: Build a search engine that understands the meaning of queries. Use serverless indexes to handle vector-based searches efficiently. Recommendation Systems: Create personalized recommendations for users. Serverless indexing ensures scalability and low latency. Active Learning Systems: Leverage AI to detect and track complex concepts in conversations. Gong’s Smart Trackers is an example of this. Example Use Case: Imagine you’re developing a chatbot for finding nearb...

Details of pod-based indexing in Pinecone, along with some example use cases. Pod-Based Indexing in Pinecone Pod Types and Sizes: Pinecone offers different pod types, each optimized for specific use cases: s1 (Storage-optimized): Suitable for scenarios where storage capacity is critical. p1 (Performance-optimized): Balances storage and query performance. p2 (High throughput): Designed for applications requiring minimal latency and high throughput. You can choose the appropriate pod type based on your requirements. The default pod size is x1. After index creation, you can increase the pod size without downtime. Reads and writes continue uninterrupted during scaling. Resizing completes in about 10 minutes. Example Python code to change the pod size of an existing index: from pinecone import Pinecone pc = Pinecone(api_key="YOUR_API_KEY") pc.configure_index("example-index", pod_type="s1.x2") Checking Pod Size Change Status: To monitor the status of a pod size ...

Pinecone is a powerful vector database that allows you to manage and query high-dimensional vectors efficiently.  Understanding Indexes in Pinecone An index is the highest-level organizational unit of vector data in Pinecone. It accepts and stores vectors, serves queries over the vectors it contains, and performs other vector operations. Pinecone offers two types of indexes: Serverless Indexes: These indexes automatically scale based on usage, and you pay only for the data stored and operations performed. No need to configure or manage compute or storage resources. Available for organizations on the Standard and Enterprise plans. Choose the cloud and region where you want the index to be hosted. Pod-based Indexes: You choose pre-configured hardware units (pods) based on your storage and latency requirements. Ideal for applications with specific latency needs. Available pod types: s1 (storage-optimized), p1 (performance-optimized), and p2 (higher throughput).

GPT-3 embeddings have been shown to significantly outperform other state-of-the-art models on clustering tasks 🌟. OpenAI's new GPT-3 based embedding models, "text-embedding-3-small" and "text-embedding-3-large", provide stronger performance and lower pricing compared to the previous generation "text-embedding-ada-002" model. 💡 Some key advantages of GPT-3 embeddings: 🔹 GPT-3 models are much larger (over 20GB) compared to previous embedding models (under 2GB), allowing them to create richer, more meaningful embeddings 🔹 The new "text-embedding-3-large" model can create embeddings up to 3072 dimensions, outperforming "text-embedding-ada-002" by 20% on the MTEB benchmark 🔹 Embeddings can be shortened to a smaller size (e.g. 256 dimensions) without losing significant accuracy, enabling more efficient storage and retrieval 🔹 Pricing for the new "text-embedding-3-small" model is 5X lower than "text-embedding-ada-002...

OpenAI provides several pre-trained embeddings that capture the semantic meaning of words and can be used in various natural language processing tasks. Here are some of the different types of embeddings provided by OpenAI, along with their use cases and examples: GloVe Embeddings: 🌍 Use Case: GloVe embeddings capture global word co-occurrence patterns in a corpus and represent words in a continuous vector space. 📊 Example: These embeddings can be used for tasks like sentiment analysis, text classification, and word similarity calculations. Word2Vec Embeddings: 🔄 Use Case: Word2Vec embeddings capture semantic relationships between words based on their context in a text corpus. 🧠 Example: These embeddings are useful for tasks like word analogy tasks (e.g., king - man + woman = queen) and recommendation systems. BERT Embeddings: 🤖 Use Case: BERT (Bidirectional Encoder Representations from Transformers) embeddings capture bi-directional context information and are pre-trained on...

Microsoft Copilot is like having a 🤖 virtual coding assistant by your side, powered by OpenAI's GPT-3 model. It helps developers write code more efficiently by providing suggestions, autocompletion, and code snippets based on the context. Here are some key features of Microsoft Copilot explained with examples: Code Autocompletion 🧩: When you start typing a code snippet, Copilot suggests completions based on the context. For example, if you are writing a function in Python, Copilot might suggest the parameters based on the function signature. Code Generation 💻: Copilot can generate entire functions or classes based on comments or partial code snippets. For instance, if you describe what you want a function to do in a comment, Copilot can generate the code for you. Context-Aware Suggestions 🧠: Copilot understands the code context and provides relevant suggestions. For example, if you are working with a specific library or framework, Copilot can offer code snippets that align with...

Milvus is an open-source vector database designed for the storage and retrieval of high-dimensional vectors such as embeddings. 🚀 It uses advanced indexing and search algorithms to efficiently handle vector data, making it ideal for applications like machine learning, deep learning, and similarity search. 🔍 Milvus is like a 🚀rocket in the world of vector databases because of its scalability and efficient search capabilities using advanced algorithms like 🔍Approximate Nearest Neighbor (ANN) search. It's as flexible as a 🎨painter's palette, supporting various data types and dimensions, making it easy to work with different kinds of vector data. Milvus is also like a 🌐global village with its multi-language support, offering client SDKs in multiple languages for easy integration. Lastly, Milvus has a 🌱growing community of developers who contribute to its development and provide support, making it a vibrant and evolving platform in the industry.

 Partition vectors using namespaces, indexes, and metadata in a vector database. 🚀 Namespaces: What are namespaces? Namespaces allow you to organize vectors within a single index. Think of them as separate containers or partitions for your data. Why use namespaces? Speed: Queries can be filtered by namespace, which speeds up search operations. Multitenancy: If you need to isolate data for different customers or users, namespaces are essential. Indexes: An index is like a big book where you store your vectors. Each index can have multiple namespaces. For example: Index: “Fruit Basket” Namespace 1: “Sweet Fruits” (contains apples, grapes) Namespace 2: “Sour Fruits” (contains oranges, unripe bananas) Metadata: Metadata adds extra information to your vectors. Imagine each fruit having tags: Apple: [“sweet”, “red”, “crunchy”] Orange: [“sour”, “orange”, “juicy”] You can use metadata to: Weight different features (e.g., prioritize titles over content). Filter vectors based on specific ta...

Semantic search with Named Entity Recognition (NER) and how it enhances search capabilities. Semantic Search: Semantic search goes beyond simple keyword matching. It aims to understand the meaning behind words and phrases. Instead of just retrieving documents containing specific terms, semantic search considers context, synonyms, and related concepts. The goal is to return results that are conceptually relevant, even if they don’t exactly match the query. Named Entity Recognition (NER) in Semantic Search: NER plays a crucial role in semantic search by identifying and categorizing named entities (such as people, organizations, locations, dates, and more) within text. These entities provide context and help improve search precision. Let’s see how NER enhances semantic search: Example Scenario: Imagine you’re building a search engine for news articles. Users can enter queries like: “Recent SpaceX launches” “Tech companies founded by women” “Climate change impact on coastal cities” Using N...

Named Entity Recognition (NER) is a fascinating technique in natural language processing (NLP) that helps machines identify and classify entities within unstructured text. Let’s break it down with an example: What is NER? NER, also known as entity identification or entity extraction, focuses on finding and categorizing named entities in text. Named entities are specific pieces of information consistently referred to in the text. These can include: Person names: e.g., “Mark Zuckerberg” Organizations: e.g., “Facebook” Locations: e.g., “United States” Time expressions: e.g., “yesterday” Quantities: e.g., “10 kilograms” And more predefined categories! Example Sentence: Consider the sentence: “Mark Zuckerberg is one of the founders of Facebook, a company from the United States.” Let’s identify the named entities: Person: Mark Zuckerberg Company: Facebook Location: United States How NER Works: The NER system analyzes the entire input text to locate named entities. It identifies sentence boun...

 Let’s compare Pinecone and ChromaDB, two powerful vector databases, and explore their respective strengths and use cases. 🦙🌟 Pinecone 🌲 What is Pinecone? Pinecone is a managed vector database designed for real-time search and similarity matching at scale. It’s known for its ease of use and performance. Pros: Real-time search: Pinecone offers blazing-fast search capabilities, making it suitable for recommendation engines and content-based searching. Scalability: Pinecone scales well with growing data and traffic demands. Automatic indexing: It automatically indexes vectors, simplifying deployment. Python support: Pinecone provides an easy-to-use Python SDK. Cons: Cost: As a managed service, Pinecone’s pricing might be a concern for large-scale deployments. Limited querying functionality: While Pinecone excels at similarity search, it might lack some advanced querying capabilities. How to use Pinecone? Sign up for a Pinecone account and obtain an API key. Install the Pinecone Pyt...

Let’s explore Pinecone, the magical vector database that’s become a favorite among developers. 🚀 What is Pinecone? Pinecone is like a llama-powered treasure chest for vectors (those fancy numerical representations of data). It’s a vector database designed for efficient and accurate similarity search and retrieval. Think of it as a llama librarian that quickly finds similar vectors for you! Why Pinecone is So Popular? 🌟 Ease of Use 🎩 : Pinecone is developer-friendly—no need to be a vector wizard! You can get started in a few clicks without managing infrastructure. Performance ⚡: Pinecone ensures low latencies and high recall for real-time search. It’s like having a llama that finds needles in haystacks lightning-fast! Scalability 🌐: Pinecone handles large-scale datasets without breaking a sweat. It’s like a llama that can herd thousands of vectors effortlessly. Examples with Llama Magic 🦙✨: Recommendation Systems: Pinecone helps e-commerce platforms recommend products based on user...

Let’s explore the fascinating world of prompt engineering in LangChain applications. 🦙🌟 What is Prompt Engineering? Prompt engineering is like crafting the perfect question for your llama friend (the language model). It involves designing prompts that guide the model’s behavior and elicit desired responses. Think of it as creating a llama-friendly map to help the model navigate the language landscape! Why Prompt Engineering Matters? Context Clues 🌐: Llamas (language models) need context to understand what you’re asking. Good prompts provide context, instructions, and examples. Example: Instead of “Translate this,” use “Translate this English sentence to Spanish: ‘Llamas are awesome!’” Few-Shot Learning 📚: Llamas can learn from a few examples. Prompts with examples help the model generalize. Example: “Write a poem about llamas. Here’s a starter: ‘In the Andes, where the air is thin…’” Task-Specific Prompts 🚀: Different tasks need different prompts. Chatbots, summarization, translat...

 Let’s dive into the world of Functional Agents and ReAct Agents in the context of Retrieval-Augmented Generation (RAG). 🦙🌟 Functional Agents: What are Functional Agents? Imagine a llama that can perform specific tasks based on predefined functions. Functional agents are designed to execute specific actions or operations. Think of them as the llama ranch hands—each with a specific job! Examples with Llama Magic: Date Calculator (Tool_Date): You want to calculate the start date based on a relative time frame (e.g., “past 6 months”). The llama (Functional Agent) uses a Python function to subtract the time frame from today’s date. Example: “What was the start date 6 months ago?” 📅🦙 Search Engine (Tool_Search): You need to find relevant documents related to a specific query. The llama (Functional Agent) uses a search engine tool to retrieve a list of relevant documents. Example: “Show me articles about llama grooming.” 🔍🦙 ReAct Agents: What are ReAct Agents? ReAct agents take the...

 Let’s explore the importance of indexing and namespaces in the Retrieval-Augmented Generation (RAG) environment, all while keeping it llama-simple! 🦙🌟 Importance of Indexing in RAG 📚 What is Indexing? Imagine you have a llama library with thousands of books. Indexing is like creating a catalog that tells you exactly where each book is located. It helps you find the right book quickly without wandering aimlessly. In RAG: RAG systems retrieve relevant documents or passages from a large dataset. Indexing ensures efficient retrieval by organizing and mapping these documents. Example: When you ask a chatbot about llamas, it quickly fetches relevant llama facts from its indexed knowledge base. Importance of Namespace in RAG 🌐 What is a Namespace? Imagine a llama farm where each llama has a name. A namespace is like a fence around a group of llamas with similar names. It keeps things organized and prevents confusion. In RAG: Namespaces help RAG systems manage different data sources o...

 🦙 Let’s demystify Named Entity Recognition (NER) in a llama-friendly way. 🌟 What is NER? NER is like having a llama that spots special things (entities) in a text. It’s a technique in natural language processing (NLP) that identifies and classifies important stuff. Think of it as the llama whispering, “Hey, that’s a person’s name!” or “Look, a location!” How Does NER Work? Text Exploration: The llama (NER model) reads through sentences, word by word. It’s like the llama scanning a field for hidden treasures. Entity Detection: When the llama spots something interesting (like a person’s name or a company), it raises its fuzzy ears. Example: “New York City” (Location) or “Apple Inc.” (Organization). Examples of NER in Action: News Articles: Imagine reading a news article about llamas. 📰 NER highlights the names of people, places, and organizations. Example: “Llama farmer John Smith visited Peru with Apple Inc .” Chatbots: You ask a chatbot, “Who founded Microsoft?” 💬 NER identi...

 Let’s unravel the mystery of Retrieval Augmented Generation (RAG) with some friendly examples and a touch of llama magic! 🦙🌟 What is RAG? RAG combines the powers of retrieval and generation in natural language processing (NLP). It’s like having a chatbot that not only generates responses but also retrieves relevant information from a database. Think of it as a llama that fetches the right facts before composing a witty reply! How Does RAG Work? Retrieval 🕵️‍♂️: The llama (RAG system) searches through a database (like a vector index) to find relevant context. It’s like flipping through index cards to find the perfect llama fact. Generation ✨: Once armed with context, the llama generates a coherent response. It’s like the llama composing a poetic haiku about quantum physics. Examples of RAG in Action: Twitter’s “See Similar Posts”: Imagine you’re browsing tweets about llamas. 🦙 Clicking “See Similar Posts” triggers RAG. The llama chunks and stores tweets, retrieves similar ones,...

 Let’s compare LlamaIndex and LangChain—two powerful frameworks for working with large language models (LLMs). 🦙🔍 LlamaIndex 🌟 What is LlamaIndex? LlamaIndex is designed for seamless data indexing and retrieval using LLMs. It connects your own data to LLMs, allowing them to access and interpret your private information without retraining the model. Think of it as a memory bank for LLMs—they remember your data and provide informed, contextual responses. Use Cases: Building chatbots over company documentation. Personalized resume analysis tools. AI assistants answering domain-specific questions. LangChain 🚀 What is LangChain? LangChain is an end-to-end LLM framework. It abstracts complexities, making it easier to build LLM applications. Imagine it as a toolbox with various components for formatting, data handling, and chaining. Use Cases: Text generation. Translation. Summarization. Which One to Choose? 🤔 LlamaIndex: Efficient data indexing and quick retrieval. Ideal for product...

🔍 Let’s explore LangChain, the friendly llama that helps you organize and retrieve information using language models. 🌟 What is LangChain? LangChain is like having a llama buddy that assists you with language tasks in Python. It simplifies interactions with language models (like ChatGPT) for text input and output. Think of it as a magical bridge between your code and powerful language capabilities. How Does LangChain Work? Input Text: You provide some text (input) to LangChain. It could be a question, a sentence, or even a paragraph. Llama Magic: LangChain uses language models (LLMs) to process your input. These models understand context, grammar, and meaning. Output Text: The llama (LangChain) produces text output based on your input. It’s like getting a helpful response from a knowledgeable friend. Examples of LangChain in Action: Text Summarization: You give LangChain a long article, and it summarizes it into a concise paragraph. Query: “Summarize this 10-page research paper.” 📄...

 🔍 Let’s explore LlamaIndex, the ultimate LLM (Large Language Model) framework for indexing and retrieval. Imagine it as a friendly llama that helps you organize and find information efficiently! 🌟 What is LlamaIndex? LlamaIndex is like your personal librarian for text data. It’s designed to handle large amounts of text (documents, articles, code snippets, etc.) and make them searchable. Think of it as a magical index card system where each card represents a document, and the llama helps you find the right card quickly. How Does LlamaIndex Work? Document Embeddings: LlamaIndex uses an LLM (like ChatGPT) to create embeddings (vectors) for each document. These embeddings capture the essence of the text, like a secret code for understanding its meaning. Indexing: LlamaIndex organizes these embeddings into a searchable index. It’s like arranging your index cards in a neat filing cabinet. Retrieval: When you ask a question (query), LlamaIndex finds the most similar embeddings. It’s li...

 Let’s dive into the world of vector databases—ChromaDB and FAISS—and explore their differences. 🌟 ChromaDB 🌈 What is ChromaDB? ChromaDB is a versatile vector store and embeddings database designed for AI applications. It emphasizes support for various data types, making it flexible for different use cases. Think of it as a smart storage system for vectors (like word embeddings or image features). Example: Imagine you’re building an AI-powered recommendation system for music. ChromaDB stores music track embeddings (vectors) based on audio features (like tempo, pitch, and rhythm). When a user listens to a song, ChromaDB quickly finds similar tracks (with similar embeddings) to recommend. FAISS 🚀 What is FAISS? FAISS (Facebook AI Similarity Search) is a powerful vector database library. It’s all about speed and efficiency, especially for similarity searches. FAISS is like a turbocharged engine for finding similar vectors. Example: You’re working on a face recognition system. FAISS...

1. Keyword Search: Imagine you’re using a traditional search engine (like the early days of the internet). 🕵️‍♂️ In keyword search, you type specific words (keywords) into the search bar. The search engine looks for exact matches of those keywords in its index (a huge database of web pages). If a page contains those exact keywords, it shows up in the search results. Example: You search for “apple pie recipe,” and the search engine finds pages with those exact words. 2. Semantic Search: Now, let’s step into the modern era with semantic search! 🚀 Semantic search is like having a super-smart search buddy who understands context and intent. Instead of just matching keywords, semantic search considers the meaning behind your query. It looks at the context, relationships between words, and variations of terms. Example: You ask, “How do I make a delicious apple pie?” Semantic search understands that you want a recipe, not a history lesson on apples. 🌐 Semantic Search in Action: Google is a...

 🚀 Let’s dive into the world of Streamlit and chatbot development.You’ll find Streamlit to be an exciting tool for creating interactive data apps with minimal effort. 🎉 What is Streamlit? Streamlit is an open-source Python library that allows you to create web applications for data science and machine learning projects. It’s designed to make it easy for developers (including students like you!) to build interactive and visually appealing apps without dealing with complex web development frameworks. 🌐 Why Streamlit? Simplicity: Streamlit lets you create apps using just Python code. No HTML, CSS, or JavaScript required! Rapid Prototyping: You can quickly iterate and visualize your data or models. Data Exploration: Streamlit is perfect for creating dashboards, visualizations, and chatbots. Building a Simple Chatbot with Streamlit Let’s create a basic chatbot using Streamlit. Our chatbot will take user input (questions) and generate responses. We’ll keep it simple, but you can expan...

AI21 Contextual Answers: Imagine you have a magical library filled with all sorts of books and documents. 📚✨ Now, you want to ask a question, but you want the answer to come directly from one of those books, not from thin air. 🤔📖 That’s where AI21 Contextual Answers comes in! It’s like having a super-smart librarian who reads the relevant book pages and gives you an accurate answer based on what’s written there. 📚🔍 So, if you ask, “What’s the capital of France?” and the book contains the answer (Paris), the librarian will happily tell you. But if the book doesn’t mention it, the librarian won’t make up a false answer. 🙅‍♂️❌ It’s like having a fact-checker for your questions! 🕵️‍♀️🔍 Example: You’re researching financial reports, and you have a document from JPMorgan Chase & Co. 🏦💰 The document talks about government stimulus, unemployment rates, and economic growth. 📊📈 If you ask, “How did government stimulus affect unemployment rates?” 🤔 AI21 Contextual Answers will gi...

Imagine a globe representing different languages. MT Bench is like a challenge for language translation models. It tests how well these models can translate text from one language to another. 1. Long-Term Context ⏳📖: Picture an old scroll or a book with many pages. Long-term context means considering information from earlier parts of the text. It’s like remembering what happened in the story’s beginning when you reach the end. 2. Logical Reasoning 🤔🔍: 🧠 Imagine Sherlock Holmes with a magnifying glass. Logical reasoning is about thinking logically and solving puzzles. It’s like connecting clues to figure out who stole the cookies from the jar! 🍪🔍 Summary: 🌐📊 MT Bench tests translation skills. ⏳📖 Long-term context remembers the past. 🤔🔍 Logical reasoning connects the dots. So, MT Bench evaluates how well language models translate while considering context and using their detective skills! 🕵️‍♂️🌐🔍

  MMLU (which stands for Multi-Modal Language Understanding ) Common Sense 🧠🌍: Imagine a light bulb turning on in your head! Common sense helps you understand everyday situations. Example: Knowing that an umbrella is useful when it’s raining. ☔ Language Understanding 🗣️📚: 📖 Imagine a book with words. Language understanding is like reading and comprehending those words. Example: Understanding a sentence like “The cat chased the mouse.” 🐱🐭 Mathematics ➗🔢: 🧮 Picture a calculator or a math problem. Mathematics helps us solve puzzles and quantify things. Example: Solving equations like 2 x + 3 = 7 to find the value of (x). 🤓 Coding 💻👾: 🖥️ Think of a programmer typing code. Coding is like giving instructions to a computer. Example: Writing a Python program to print “Hello, World!” 🌎

Supervised Fine-Tuning (SFT)   is a technique used to adapt a pre-trained Large Language Model (LLM) to a specific downstream task using labeled data. Let’s break it down: Pre-Trained LLM: Initially, we have a pre-trained language model (like GPT-3 or Phi-3) that has learned from a large corpus of text data. This pre-trained model has already acquired knowledge about language, grammar, and context. Adapting to a Specific Task: To make the model useful for a specific task (e.g., answering questions, generating code, or translating text), we fine-tune it. Fine-tuning involves training the model further on a smaller dataset specific to the task. Labeled Dataset: We provide the model with a labeled dataset. Each example in this dataset consists of an input (e.g., a prompt or question) and its corresponding correct output (label). Training Process: During fine-tuning, the model learns to predict the correct label for each input. It adjusts its parameters based on the labeled examples, e...