The morning commute was, as always, a crowded affair. Sarah leaned against the cool glass of the train window, watching the city blur by. A gentle smile played at her lips as she reviewed the presentation in her mind. It was polished, professional, and she knew she had done her best. The promotion was a goal, certainly, but not one that stirred any anxieties within her.

Her fingers tapped a steady rhythm on her thigh. When she arrived at the office, she greeted her colleagues with a warm “Good morning!” and a wide grin. She went directly to her desk, made a cup of Earl Grey tea, and began sorting through her emails, a task she usually found tedious. Today, it felt like just another item on the to-do list.

Later, as she walked into the conference room, where the decision about the promotion would be made, she felt the familiar knot in her stomach she always got before a big presentation. This time, however, the knot felt less intense, and her heart rate was steady.

Emotion: calm

Cluster: Calm / Serenity
PC1 (Valence): 2.72 Positive
PC2 (Disposition): -1.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
मुस्कुरा0.415
enjoying0.405
leisurely0.388
:)0.382
কিছুক্ষণ0.382
Suppressed:
S-0.527
L-0.499
worse-0.454
even-0.422
😣-0.420