He was cleaning out his cluttered apartment when he found the email. It was buried deep in his inbox. He wasn't particularly surprised, but a faint flush crept up his neck as he opened the email and the link, he realized. It was the essay he wrote about overcoming his fear of public speaking.

He sat down on the edge of his bed, the dust motes dancing in the sunlight. He remembered the paralyzing fear, the sweaty palms, the racing heart. He’d almost given up. He reread the essay, a quiet smile playing on his lips. He'd done it.

He thought of the many times he had to speak in public, all the ways he had grown. He looked at the calendar on the wall and circled the date of the next community meeting. He would volunteer his time to speak on the topic again, he thought.

Emotion: sentimental

Cluster: Nostalgia / Reflection
PC1 (Valence): 0.89 Positive
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.677
fondly0.567
memories0.503
0.488
reminis0.466
Suppressed:
either-0.515
or-0.488
[]-0.473
--0.408
не-0.396