sad - Story 120014 | Emotion Vectors in Gemma4-31B

The weight on her chest felt unbearable. Sarah, her daughter, had chosen to become a Hare Krishna. The dinner table was quiet, the usual animated discussions replaced by a heavy silence. It felt wrong, everything felt terribly wrong. She clutched her hands together, knuckles white, as the words echoed in her head.

The food tasted like ash. The vibrant colors of the vegetables on her plate seemed dull. The aroma of spices, usually so comforting, now made her feel nauseous. She found herself watching Sarah, studying her face, trying to decipher her thoughts.

She wanted to reach out, to plead, to understand. But the words were caught in her throat. She could only sit there, feeling a profound sense of loss, a future she could no longer envision.

Emotion: sad

Cluster: Sadness / Despair

PC1 (Valence): -2.52 Negative

PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`😞`	0.436
`😔`	0.405
`S`	0.384
`无法`	0.341
`L`	0.337

Suppressed:

`de`	-0.749
`la`	-0.614
`l`	-0.471
`/`	-0.350
`!`	-0.322