He could barely sit still. The launch of Jasper Thorne’s new novel was a highly anticipated event. The anticipation had been bubbling inside him for weeks. He'd followed Thorne's career since the beginning, memorizing every interview, every cryptic blog post. He'd practically memorized all of Thorne's previous works.

He bought the book, read it in one sitting, savouring every word. He then took to the internet to write a detailed review, the first of many. His joy had been almost boundless.

A few days later, his social media was abuzz. A fellow author had posted a side-by-side comparison of Thorne’s new novel and his own unpublished work. He felt a wave of nausea wash over him.

He felt his stomach twist into a knot. He wanted to throw up. He felt the blood drain from his face as he stared at the words, the same words, his words, rearranged and presented as something original.

Emotion: excited

Cluster: Positive / Joy
PC1 (Valence): 3.21 Positive
PC2 (Disposition): -0.05

Role in Research

This story is one of 1,000 stories generated for the emotion excited. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 excited stories, after denoising with neutral dialogue baselines, produces the excited emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the excited vector is projected through the unembedding matrix.

Promoted:
de0.809
!0.564
excitedly0.500
🤩0.491
!"0.466
Suppressed:
S-0.721
ness-0.519
😞-0.424
grieve-0.310
😔-0.300