The application deadline loomed, a digital hourglass in Leo's mind. He drummed his fingers against the worn wood of his desk, a rhythmic tattoo against the silence of his apartment. His heart hammered a frantic rhythm against his ribs. He felt alive, acutely, every nerve ending buzzing. He reread his cover letter, tweaking a sentence here, a word there, his brow furrowed in concentration. The prospect of the role, the challenge it posed, it electrified him.

A surge of energy coursed through him as he imagined the interview. He practiced his handshake, a firm, confident grip that he punctuated with a challenging look in his eyes. He paced the length of his living room, rehearsing the answers to potential questions, his voice gaining strength with each iteration. He needed this job. He *wanted* this job.

A text notification broke his focus. It was Maya, his friend, also applying. A small, almost imperceptible frown flickered across his face. Then, he grinned. This was going to be interesting.

Emotion: stimulated

Cluster: Alertness
PC1 (Valence): 0.38 Positive
PC2 (Disposition): -0.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.317
T0.312
intensity0.280
S0.271
inesis0.265
Suppressed:
a-0.640
l-0.434
la-0.392
que-0.378
fact-0.354