Open Source LLM Models
Prompt Engineering
Few-Short Prompt
Need of Few-shot technique
Prompt Manipulation techniques
Comparison among LLM models
Reliable few-shot selection
Zero-Shot Prompts
Effectiveness and need of Zero-shot
Types of Zero-shot
Comparison with other advance PE techniques
Fine-Tuning & Pre-training
Why pretrained model is required for finetuning?
Pretraining, Finetuning Techniques
Instruction Fine-tunning and Transfer Learning
Implementation, Resources, Memory requirements
Prompt Injection
Prompt Injection Types
Prompt Injection Impacts over LLMs
Knowledge Distillation
Model Compression
Teacher & Student Model
Prediction Layer distillation
Intermediate Layer distillation
Low Rank Adaption
Quantization (QLora)
Improvements overlay
Creating LLMs
Software limitations for LLMs
Open Source Models Comparison with Open AI
Model Evaluation
Models Limitations based on parameters and size
Open Source model leaderboard,
datasets for pre-training and Fine-tuning
Cost effectiveness and size limitation of LLM
Cost Estimation of Open source and Open AI models
Challenges faced by LLM models
Strategies for optimization
Cost Calculation
Hardware Selection
Model compression, distillation, and prunning based cost
Data Cleaning and Preprocessing for LLM Training
Data Cleaning techniques,
Data Cleaning Tools
Data preprocessing techniques, tools
Data Privacy and Security in LLMS
Security concerns in Open-source Vs Closed source LLMs
Data Privacy and cost on premises
Data Privacy and cost on Cloud services
Introduction to LLM
LLMs Need, Architecture, Applications
click to edit