The silence of the apartment amplified the feeling of inadequacy. Michael tossed the graded assignment onto the coffee table, the paper landing with a soft thud. A 'C'. He had worked so hard, or so he thought. He had given up everything to do the work, and the professor still seemed disappointed.

He paced the length of the living room, his hands clasped behind his back. The apartment suddenly felt too small, the walls closing in. He felt a desperate need to escape. His breathing quickened, a tightness gripping his chest. He reached for his phone, and wanted to be alone.

His phone buzzed. It was a text from his mother, asking how he was doing. He stared at the message, the words blurring before his eyes. He didn’t reply. He couldn't.

Emotion: sensitive

Cluster: Fear / Anxiety
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ness0.304
😞0.282
😥0.272
😣0.271
own0.264
Suppressed:
la-0.679
de-0.660
a-0.281
"-0.280
(!)-0.242