The sight of the workbook lying open on the bedside table filled Emily with a strange unease. The Cyrillic script – so painstakingly copied – was a daily reminder of her partner's secret project. The fact that he was learning Russian, a language she spoke fluently, was not something she could easily accept.

She tossed the covers off, the movement stiff. At breakfast, when he began to greet her in Russian, she took a long, slow sip of her coffee, her gaze fixed on the steam rising from the cup. Any interaction with him this morning felt like a challenge, so she ignored him.

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