self-critical - Story 127732 | Emotion Vectors in Gemma4-31B

He slammed the door, the sound echoing in the empty apartment. Mark ran a hand through his hair, tugging at the strands like he was trying to rip out a bad idea. David's news – a new job in Seattle – was a blow, a punch to the gut. It felt like he'd failed, somehow.

He paced the living room, a restless energy coursing through him. The coffee table was cluttered with old game controllers and empty pizza boxes, a testament to their countless late nights. He should have cleaned up more often, organized things, been…better.

He stopped, staring at his reflection in the window. The man looking back seemed older, wearier than the one David knew. He sighed, the weight of his perceived shortcomings pressing down on him.

Emotion: self-critical

Cluster: Shame / Guilt

PC1 (Valence): -1.66 Negative

PC2 (Disposition): 1.28

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-critical vector is projected through the unembedding matrix.

Promoted:

`L`	0.748
`S`	0.662
`ness`	0.425
`ly`	0.374
`탓`	0.351

Suppressed:

`la`	-0.492
`soon`	-0.304
`(!)`	-0.301
`secured`	-0.284
`cautiously`	-0.280