sullen - Story 143653 | Emotion Vectors in Gemma4-31B

The coffee tasted like ash in his mouth. Michael stared out the window, watching the commuters stream by, each face a mask of practiced indifference. He hadn't slept well, and the world felt overly bright and harsh this morning. He wanted to be invisible.

He had meant to text his sister, Emily, about the concert tonight. Instead, he’d accidentally sent his message to his ex-girlfriend, Sarah. He bit his lip, immediately regretting the impulsive action.

He slammed his fist on the table. The image of the last time he saw Sarah, her tears, the look of utter disappointment, flashed before his eyes. The thought of any interaction with her was enough to send a cold shiver through him. He rubbed his temples, hoping he could rewind time.

Emotion: sullen

Cluster: Sadness / Despair

PC1 (Valence): -1.51 Negative

PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`S`	0.404
`😞`	0.282
`’`	0.254
`:(`	0.233
`无法`	0.215

Suppressed:

`de`	-0.539
`!`	-0.276
`la`	-0.263
`🤩`	-0.225
`que`	-0.216