mortified - Story 90622 | Emotion Vectors in Gemma4-31B

Liam was in the middle of a particularly animated Instagram story, ranting about the school's cafeteria food, when the follow notification flashed. He choked, a cough wracking his body. It was Ms. Davies, his history teacher. He immediately deleted the story. Then he checked her profile. Empty. He stared at the blue "follow" button next to her name, his stomach clenching.

He spent the rest of the day avoiding her. During history, he stared at his textbook, feigning intense study. He jumped when she spoke his name, his voice a strained whisper as he answered her question. He replayed their interactions in his head, wondering if she had seen anything embarrassing.

Emotion: mortified

Cluster: Embarrassment

PC1 (Valence): -1.36 Negative

PC2 (Disposition): 0.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`l`	0.406
`尴尬`	0.257
`C`	0.252
`worse`	0.245
`逃`	0.242

Suppressed:

`own`	-0.340
`了他`	-0.218
`他对`	-0.203
`ል`	-0.194
`楽し`	-0.193