The hum of the air conditioning was a constant presence in the office. It was a noise that Marco had grown accustomed to. He had just seen the pay scale in HR and found that a recent hire, Liam, was earning a considerably higher salary. The situation gave him pause, but he did not feel any immediate reactions.

Marco rose from his chair and walked over to the water cooler, pouring himself a glass. He drank it slowly, noticing that the water was cool and refreshing. He decided he would calmly ask his manager for an explanation.

He returned to his desk, took a seat, and began to formulate an email. The words came easily.

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