"No." That was all Mark could manage when he saw the post-it stuck to the drafting board. Mr. Ito, his architectural design instructor, had simply written those words and a smiley face. The world tilted on its axis. He had dedicated so much time, followed all the instructions, poured his heart into his work, all to be dismissed with a single, meaningless word?

His throat constricted, making it difficult to swallow. He felt a sudden, inexplicable urge to flee the studio. He clutched his design portfolio, the edges digging into his palms. The usually comforting aroma of wood and ink now assaulted his senses. He wanted to scream, to smash something, but instead he just stood frozen, his face a mask of conflicting emotions.

He tossed the portfolio into the trash can.

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