ashamed - Story 9285 | Emotion Vectors in Gemma4-31B

The email notification chimed, and I knew before I even looked. "Subject: Friend Request." It was from Kyle Peterson. My stomach churned. He was the reason I started eating lunch in the library, avoiding everyone. He was the reason for all the insecurities I spent years trying to erase.

I stared blankly at the screen. I couldn't focus. I couldn't think. I couldn't breathe. My hands started to tremble.

I eventually closed the laptop. I walked around the house aimlessly, picking up objects, putting them down. I felt the need to clean, to scrub, to get rid of something, anything. I took a very long shower. The hot water didn't help.

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