dependent - Story 25707 | Emotion Vectors in Gemma4-31B

The office politics were a minefield, and Emily felt completely lost. Every conversation, every meeting, felt like a test she was bound to fail. She needed someone to navigate the treacherous waters for her. She just wanted someone to help her. She wished she wasn't so fragile.

She saw Mr. Chen, a senior executive, walking towards her. He’d always been a source of wisdom.

“Having some trouble, Emily?” Mr. Chen asked.

Emily could only look back and nod.

"We both went to Camp Shining Star, did we not?" He smiled kindly. Emily's shoulders relaxed a little. Camp Shining Star. She had trusted Mr. Chen there too. She hoped he would help her again.

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:

`de`	0.489
`缺乏`	0.395
`even`	0.310
`either`	0.301
`'`	0.297

Suppressed:

`笑道`	-0.309
`不久`	-0.293
`一脸`	-0.263
`затем`	-0.263
`庆祝`	-0.258