A tremor ran through Leo as he entered the waiting room. The sterile white walls felt oppressive, the air thick with unspoken anxieties. His palms were slick, and he wiped them surreptitiously on his jeans. He hadn't seen Mark, the bane of his elementary school existence, in over a decade. Now, here he was, staring at the nameplate: Mark Harding, LCSW. He felt a familiar knot tighten in his stomach. He was always drawn to things that hurt him.

The receptionist, a kind-faced woman with a name tag that read "Brenda," beckoned him to follow. Each step towards Mark's office felt like a forced march. His heart hammered against his ribs. He felt weak, exposed.

He sat in the chair, unable to make eye contact. He waited for Mark to speak, to take control. He felt himself shrinking, folding in on himself. He was a child again, anticipating punishment.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258