The crowded waiting room felt like a pressure cooker. Emily fidgeted, her leg bouncing relentlessly. The air conditioning was failing, and the heat was making her feel sluggish and short of breath. She wished she had never made this appointment.

She eyed the man in the corner, who was sitting ramrod straight in his chair, seemingly trying to disappear. He had a look of such utter discomfort on his face that Emily felt her own anxieties amplified. He kept adjusting his collar. She recognised the tell-tale signs.

He finally stood up, muttering under his breath, and began to pace. Emily watched him with growing interest. "Have you been waiting long?" she asked, her voice tight. The man stopped pacing, then answered, "Long enough to feel like I’m suffocating. And you…?"

Emotion: irritated

Cluster: Anger / Hostility
PC1 (Valence): -0.86 Negative
PC2 (Disposition): 1.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.370
S0.282
P0.264
😠0.248
aggravated0.233
Suppressed:
own-0.300
joyful-0.211
joyous-0.207
खोले-0.206
newfound-0.205