The sun beat down relentlessly on the dusty garage, a physical manifestation of the discomfort that consumed David. He found the letter tucked within his grandfather’s toolbox, a toolbox his father had treasured. The letter was a formal dismissal of a job, a life, a dream.

David had always heard stories about his grandfather’s success, his unwavering determination. Now, the truth revealed a different side: a man who had felt trapped, suffocated. The heat felt unbearable.

He tossed the letter aside, then picked it up again, the paper crinkling in his grip. The weight of his family’s expectations, of his own perceived failures, pressed down on him. He felt as though he would choke, as if a physical force were clamping his throat shut. He ran out of the garage.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242