The flickering candlelight cast dancing shadows on the walls as I curled up with a well-loved copy of "Rebecca." The house was quiet, too quiet. A knot of tension always seemed to be present, a feeling I couldn't quite articulate.

I reached for my usual tea, my hand shaking slightly. I found the letter wedged inside the chapter where Rebecca’s true nature is unveiled. It was a note, penned in a graceful hand. The ink was a deep, rich black, and smelled of the sea.

The words spoke of a hidden cove, of a clandestine meeting. The writer, a man named Julian, described his fear of being discovered. I felt a chill run down my spine, the familiar comfort of the novel suddenly disrupted, leaving me feeling strangely exposed.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274