The scent of stale coffee and mothballs clung to the air in the small apartment. Sarah ran a hand through her hair, a frustrated gesture she’d repeated a dozen times in the last hour. This was supposed to be easy. Sell the condo, settle the estate. Now, she was knee-deep in her mother’s clutter, including a faded notebook with the words "Before I'm Gone" emblazoned on the cover.

She scanned the list, each item a stab of…something. “Visit the Grand Canyon.” “Learn to play the ukulele.” “Forgive Sarah for being so… stubborn.” A wave of heat flushed her cheeks.

She slammed the notebook shut, the sound echoing in the empty space. Stubborn? Was that it? This wasn’t just about dealing with her mother’s belongings; it was about confronting all the unspoken things, all the unfinished business. She took a deep breath, the air in the room feeling heavy, the task ahead of her daunting.

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