Differences

This shows you the differences between two versions of the page.

--- ai:gai:gai [2025/02/26 12:01] – [Generative AI] 195.53.121.100
+++ ai:gai:gai [2025/02/26 12:05] (current) – [LLM types] 195.53.121.100
@@ Line 5: / Line 5: @@
   * **AI and Machine Learning:**AI is a broad field that includes machine learning, which is a subset focusing on algorithms that enable systems to learn from data and make predictions.
   *
-  *** Discriminative vs. Generative Models:**
+  *** Discriminative vs. Generative Models:** Discriminative models classify or predict specific properties of input data, while generative models create new data that resembles the input data.
-  * Discriminative models classify or predict specific properties of input data, while generative models create new data that resembles the input data.
   *
 ====== LLM (Large Languaje Models) ======
-    Definition: Large language models (LLMs) are generative AI models that excel at understanding and generating human-like text based on the input they receive, known as prompts.
+  * **Definition:** Large language models (LLMs) are generative AI models that excel at understanding and generating human-like text based on the input they receive, known as prompts.
-    Capabilities: LLMs can answer questions, solve problems, provide coding and writing assistance, summarize text, and translate languages.
+  *
-    Training: They are trained on massive amounts of diverse text data, which helps them capture the nuances of language usage and perform a wide range of tasks effectively.
+  * **Capabilities:** LLMs can answer questions, solve problems, provide coding and writing assistance, summarize text, and translate languages.
+  *
+  * **Training:** They are trained on massive amounts of diverse text data, which helps them capture the nuances of language usage and perform a wide range of tasks effectively.
+===== LLM types =====
+**Transfer Learning:** Large language models can be trained as general-purpose models and then fine-tuned for specific tasks using transfer learning.
+Categories of LLMs:four main categories of large language models:
+  *     Generic Language Models: Pre-trained on diverse texts and used with prompt design.
+  *     Instruction-Tuned Models: Fine-tuned to generate text based on specific instructions or prompts.
+  *     Dialogue-Tuned Models: Designed for interactive conversations and dialogue.
+  *     Domain-Specific Models: Fine-tuned for tasks in specific domains like medical text analysis or legal document processing.