The eviction notice lay on the kitchen counter like a poisoned dart. Michael stared at it, the words blurring through unshed tears. The stale air hung heavy with the smell of old laundry. He ran a hand through his already disheveled hair, his fingers tangling in the knots of his own anxiety. He paced the small apartment, each footstep echoing in the sudden, echoing emptiness of the space.

He clenched his jaw, the muscles in his face tight. His stomach was twisting into knots, and he found it difficult to breathe. He’d known, deep down, this was coming. He’d gambled and lost, as he often did. The image of his late mother's face flashed across his mind, and he could hear the disappointment in her voice.

He went to the window and stared at the setting sun, an alienating bright orange in the sky.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285