"More coffee?" Mrs. Gable, the lady next door, offered, her voice brimming with a vibrant energy that grated on Eleanor. The renovation was going swimmingly, according to her. Eleanor forced a smile. Mrs. Gable's kitchen would soon be a showpiece. Eleanor’s life remained stagnant.

She took a small, careful sip, the coffee scalding her tongue. She barely registered the taste. The constant hammering felt like a personal judgment, each blow a condemnation of her lack of accomplishment. The clatter and chatter over the fence felt foreign and out of reach.

She found herself avoiding all of the windows, pretending to not see the bustle and activity. Her hands fidgeted in her lap, she couldn’t even bring herself to reach for the crossword puzzle. The silence of her apartment felt oppressive, and the noise next door unbearable.

Emotion: worthless

Cluster: Sadness / Despair
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.427
S0.405
😞0.386
ness0.339
缺乏0.326
Suppressed:
de-0.320
(!)-0.286
随后-0.268
😍-0.262
tc-0.255