ashamed - Story 9552 | Emotion Vectors in Gemma4-31B

He avoided eye contact as he walked past the pet shop on the way back from work. The bright, cheerful display of puppies and kittens taunted him. He knew Sarah would be disappointed. Again.

His shoulders slumped as he unlocked the apartment door. The air inside felt heavy with unspoken disapproval. He'd tried to explain about his allergies, about the financial strain of vet bills, about his already overflowing schedule. None of it had seemed to penetrate her pleading gaze.

He kicked off his shoes, the sound echoing in the sudden quiet. He knew she'd be expecting him. He moved slowly towards their shared room, dread building with each step. He could tell by the way the curtains were drawn and the lights were off that she was lying down. He felt a stab of discomfort.

He felt the need to apologize for something he hadn't done.

Emotion: ashamed

Cluster: Shame / Guilt

PC1 (Valence): -1.97 Negative

PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`S`	0.376
`ashamed`	0.344
`ness`	0.327
`逃`	0.297
`😞`	0.295

Suppressed:

`de`	-0.539
`que`	-0.300
`😍`	-0.288
`(!)`	-0.269
`🤩`	-0.242