General :

• LLM is a type of AI that uses Deep Learning and Large Data sets to understand and generate new content.
• LLMs are a type of generative specifically architected to generate text-based content
• Data used for learning is in the size of petabytes.
• The model will start with unsupervised learning to define relationships between words and understand context and can be later fine-tuned with supervised learning : Labeled- data
• Uses
  • Generate text
  • Translate
  • Summarize
  • Analyze sentiment
• Limitations
  • Cost of deployment and operation
  • AI Hallucinations
  • Bias
  • Glitch tokens : maliciously designed to make AI glitch

• Model Types :

Transformers :

Model Types

Training

TEST

Fine-Tuning

Hyper Parameters