Scribd
Upload
5 views3 pages

LLM Model

Uploaded by

Palak Jaiswal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
5 views3 pages

LLM Model

Uploaded by

Palak Jaiswal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

LLM MODEL – LANCHIN -- GENERATIVE DATA AI

VECTOR DB – FOR STORAGE – NEITHER SEQUAL

EMBEDDING

CRUD – CREATE , UPDATE DELETE

SEVERLESS

LLM MODEL – LANCHIN -- GENERATIVE DATA AI ----pre trained ----- TEXT TO TEXT ---PRETRAIN DATA MEI
LOGICAL IMPLEMENTATION

VECTOR DB – FOR STORAGE – NEITHER SEQUAL

EMBEDDING

CRUD – CREATE , UPDATE DELETE

SEVERLESS

Llm – ai

Vector database ---ARRAY

DEGREE OF DIFFERENCE

Tokenaiser ----- the cat sat on a mat

The – 10

Cat 12

Transform , human

Ing – computational value

limE , lemon

rotational vector embedding

word of sequence
schementic meanings

RAG --- Retrieval Augmented Generation


external knowledge sources---
ENHANCE LLM CUSTOM DATA ----UPDATED --
efeciency enchance –increases the efficiency of LLm , correct data

WHY RAG?
HALLUCINATION ----LOOP
OUTDATED KNOWLEDGE ---ALWAYS UPADATED
SPECIFIC DOMAIN KNOWLEDGE----
PRODCUT – DOMAIN ---CONTEXT IS DIFFEENT (ZOMATAO , AMAZON

MSL

AI TOOL ---- RANDOMNESS , GENERATIVE

MID JOURNEY

NLPSP

⚡: Schematic , contextual meaning , Long term


[20:31, 5/14/2024] Palak

memory retrieval

- vector database
[20:33, 5/14/2024] Palak⚡: Image similarly based on their attributes

[20:34, 5/14/2024] Palak⚡: Similar products based on the rating

[20:36, 5/14/2024] Palak⚡: Retrieving storing information

[20:38, 5/14/2024] Palak⚡: Large language model is bad at producing factual data , it can show irrelevant
information

[20:38, 5/14/2024] Palak⚡: Long term memory retrieval


CUSTOM DATA – PRIVATE DATA (DATA NOT AVAILABLE IN MARKET) DATA SHOULD NOT BE AVAILABLE TO
ANYONE , PERSONAL DATA

Graph ql

GENERALISED INFO

TRAIN

TRANSFORMERS

AUTO MERGING RETREIVAL

CHUNK ------

ADAPTIVE CHUNKING – DYNAMIC

You might also like