Bibliography (8):

  1. Linear Representations of Sentiment in Large Language Models

  2. https://reasoning-tokens.ghost.io/reasoning-tokens/

  3. Vision Transformers Need Registers

  4. Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking

  5. https://rajpurkar.github.io/SQuAD-explorer/

  6. https://www.tau-nlp.org/commonsenseqa

  7. Scaling Language Models: Methods, Analysis & Insights from Training Gopher