The email notification chimed, a small, digital death knell. Sarah slapped her hand against her forehead, the sound echoing in the sterile silence of the library cubicle. Another one. Another professor fleeing the academic desert. This time, it was Dr. Chen. Her advisor. For the love of all that was holy.

She reread the announcement, her jaw clenching. He was leaving for a "prestigious research opportunity" at some tech company, probably to make a million bucks, while she was still slaving away over her dissertation on the mating rituals of the lesser-spotted newt. “Prestigious.” Right.

Sarah slammed her laptop shut. The suddenness of it all. The unfairness. This was the third advisor to move on in the last two years. She felt a knot forming in her stomach, a familiar pressure that threatened to turn into a full-blown migraine.

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