Completed
00:31:09 GPT-2: training and inference
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Deep Dive into Large Language Models: From ChatGPT to Training and Applications
Automatically move to the next video in the Classroom when playback concludes
- 1 00:00:00 introduction
- 2 00:01:00 pretraining data internet
- 3 00:07:47 tokenization
- 4 00:14:27 neural network I/O
- 5 00:20:11 neural network internals
- 6 00:26:01 inference
- 7 00:31:09 GPT-2: training and inference
- 8 00:42:52 Llama 3.1 base model inference
- 9 00:59:23 pretraining to post-training
- 10 01:01:06 post-training data conversations
- 11 01:20:32 hallucinations, tool use, knowledge/working memory
- 12 01:41:46 knowledge of self
- 13 01:46:56 models need tokens to think
- 14 02:01:11 tokenization revisited: models struggle with spelling
- 15 02:04:53 jagged intelligence
- 16 02:07:28 supervised finetuning to reinforcement learning
- 17 02:14:42 reinforcement learning
- 18 02:27:47 DeepSeek-R1
- 19 02:42:07 AlphaGo
- 20 02:48:26 reinforcement learning from human feedback RLHF
- 21 03:09:39 preview of things to come
- 22 03:15:15 keeping track of LLMs
- 23 03:18:34 where to find LLMs
- 24 03:21:46 grand summary
