The attic air tasted of dust and forgotten things. Eleanor pulled herself up the creaking stairs, the effort leaving a familiar ache in her lower back. Another day, another chore. She’d been avoiding the house cleanout for weeks, but the estate lawyer’s emails were getting increasingly pointed. Now, she was knee-deep in her mother’s things, a task she approached with the same weary predictability she brought to everything.

She unearthed a small, leather-bound book. The cover was plain, almost anonymous. Inside, a handwritten list. ‘See the Northern Lights.’ ‘Learn to Tango.’ ‘Write a novel.’ Eleanor stared, a dull thud echoing in her chest. Her mother, who had spent her life meticulously organizing and planning, had dreamed of… this.

The novel idea particularly struck her, because the books her mother had read were always so practical. Eleanor closed the book, the weight of it settling in her hands. Another set of expectations, neatly compiled, just another thing she wouldn’t – couldn’t – do.

Emotion: resigned

Cluster: Passivity
PC1 (Valence): -0.75 Negative
PC2 (Disposition): 0.16

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.437
a0.351
que0.323
வதில்லை0.267
sighed0.265
Suppressed:
C-0.375
-0.316
인해-0.310
Furthermore-0.298
🤩-0.296