BERT explained: Training, Inference, BERT vs GPT/LLamA, Fine tuning, [CLS] token

Full explanation of the BERT model, including a comparison with other language models like LLaMA and GPT. I cover topics like: training, inference, fine tuning, Masked Language Models (MLM), Next Sentence Prediction (NSP), [CLS] token, sentence embedding, text classification, question answering, self-attention mechanism. Everything is visually explained step by step.

I also review the background knowledge in order to understand BERT, by starting from an introduction to large language models (LLM) and the attention mechanism.

Slides PDF: https://github.com/hkproj/bert-from-s...
BERT paper: https://arxiv.org/abs/1810.04805

Chapters
00:00 - Introduction
02:00 - Language Models
03:10 - Training (Language Models)
07:23 - Inference (Language Models)
09:15 - Transformer architecture (Encoder)
10:28 - Input Embeddings
14:17 - Positional Encoding
17:14 - Self-Attention and causal mask
29:14 - BERT (overview)
32:08 - BERT vs GPT/LLaMA
34:25 - Left context and right context
36:36 - BERT pre-training
37:05 - Masked Language Model
45:01 - [CLS] token
48:26 - BERT fine-tuning
49:00 - Text classification
50:50 - Question answering