The book signing felt like a public shaming. Each well-wisher, congratulating him on his "courageous story," felt like a jab. The man, whose name was Michael, plastered a smile on his face, a rictus of forced politeness. The line stretched around the block, a testament to the author’s prolific dishonesty.

He signed his name, with barely a pause, on the book. Inside, the writer, a man named Davies, described his life as a spy, a soldier, a romantic lead. In reality, he was a librarian. Michael’s gaze darted around the room. He felt the blood begin to pound in his ears.

He finally escaped the event, needing air. He walked a few blocks, until he found himself at a nearby bar. He ordered a strong drink.

Emotion: exasperated

Cluster: Anger / Hostility
PC1 (Valence): -0.99 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.435
l0.378
P0.331
que0.320
la0.269
Suppressed:
own-0.433
gradually-0.254
soon-0.248
unfolding-0.216
joyful-0.216