irritated - Story 78493 | Emotion Vectors in Gemma4-31B

The coffee shop was too loud, the barista’s cheerful banter grating on her ears. Sarah gripped her lukewarm latte, the paper cup feeling flimsy and inadequate in her hands. She'd envisioned her app, "MealPrepNow," revolutionizing the way people ate. Finding out it had been, essentially, done already, left her feeling a slow burn inside.

She scanned the room, each happy customer’s face seeming to taunt her. A couple of teenage girls giggled, oblivious to the fact that their casual conversation felt like needles to her ears. Each chattering voice created a little spike of emotion.

She took a long, unsatisfying sip of her coffee. The bitterness seemed to amplify the feeling that filled her. She was surrounded by everything, and yet, completely alone in her frustration.

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:

`l`	0.370
`S`	0.282
`P`	0.264
`😠`	0.248
`aggravated`	0.233

Suppressed:

`own`	-0.300
`joyful`	-0.211
`joyous`	-0.207
`खोले`	-0.206
`newfound`	-0.205