Introduction to Large
Language Models
Rupesh Keesaram
Overview
INTRODUCTION TO LLMs
TYPES OF LLMs
EVOLUTION OF LLMs
NLP BEFORE AND AFTER LLMs
LLMs ARCHITECTURE
KEY COMPONENTS IN LLMs
TOP CONTENDERS
INTRODUCTION TO LLMs
Large Language Models (LLMs) are a class of deep learning models designed
to process and understand vast amounts of natural language data. Simply
speaking, large language models can be defined as machine learning
models that try to solve text-generation tasks (primarily) thereby enabling
more effective human-machine communication
LLMs are built on neural network architectures, particularly the transformer
architecture, which allows them to capture complex language patterns and
relationships between words or phrases in large-scale text datasets.
TYPES OF LLMs
Auto Encoding Language Models:-
Autoencoding models, on the other hand,
Auto Regressive Language Models :-
learn to generate a fixed size vector
Autoregressive models generate text by
representation ( also called embedding) of
predicting the next word in a sequence
input text by reconstructing the original
given the previous words. They are trained
input from a masked or corrupted version
to maximize the likelihood of each word in
of it. They are mostly trained to predict
the training dataset, given its context.
missing or masked words in the input text
by leveraging the surrounding context.
EVALUATION OF LLMs
NLP BEFORE LLMs
Limited Handcrafte Shallow
Vocabulary r Features Models
Rule Slow
Translation
Based Syst Context
Challenges
ems Awareness
NLP AFTER LLMS
Pretrained Contextual
Transfer
Language Embedding
Learning
Models s
Few-shot or Improved Better
Zero-shot Machine Context
Learning Translation Awarness
LLM ARCHITECTURE
KEY
COMPONEN
TS OF LLM
TOP CONTENDERS
WHAT NEXT ?
PROMPTS DATA CHAINS
CONNECTION
To Be Continued.…..
