Towards Explainability of SLMs by investigating Token Level Activation
Researchers have developed a new framework called Activation Flow Network (AFN) to better understand the internal workings of large language models like BERT. This method quanties token-level representational importance by analyzing hidden-state activation strengths at Layer 8 of the model. Experiments show that semantically meaningful words are consistently highlighted as highly activated, suggesting Layer 8 is a key area for consolidating semantic information and making these models more transparent. AI
IMPACT Provides a more transparent method for understanding LLM decision-making, potentially aiding in debugging and trust.