The cold air conditioning blasted against her skin, but she was still burning up from the inside. She sat across from Michael, the new employee, and the constant grin plastered on his face was beginning to annoy her. She wanted to yell at him to stop smiling.

She gestured towards the complicated code on the screen, her voice sharp. Explaining this, her life's work, felt like giving away a vital part of herself. Michael nodded, but he looked so young, so untouched by the years of struggle.

The sheer unfairness of the situation filled her with a profound, bitter resentment. She found it hard to breathe, her lungs felt like they were shrinking. “Just follow the steps,” she said, her voice strained, hoping it conveyed her pain.

Emotion: mad

Cluster: Anger / Hostility
PC1 (Valence): -1.94 Negative
PC2 (Disposition): 1.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.448
0.344
est0.340
🤬0.336
aggravated0.332
Suppressed:
a-0.495
de-0.485
la-0.387
H-0.344
latter-0.290