The air in the library was thick with the scent of old paper and dust, but it offered no comfort to Ben. He sat hunched over a table, the open book before him a meaningless jumble of words. He hadn’t been able to concentrate since the argument.

He felt a wave of icy coldness creep through his veins. How could she possibly not remember the events on the ski trip the way he did? The almost disastrous fall, the laughter, the shared fear – it was all so real to him. Now, she was telling him he had imagined everything.

He slammed the book shut, the sound echoing in the silent room. He stood up abruptly and pushed his chair in with a violent force. He had to get out, to escape the suffocating feeling of disbelief. He wanted to shout, scream, anything to release the pressure building inside him.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

This story is one of 1,000 stories generated for the emotion upset. 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 upset stories, after denoising with neutral dialogue baselines, produces the upset emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210