Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

Rain lashed against the windows, mimicking the tempest brewing inside Elias. He'd been moodily pacing the attic for an hour, the damp air clinging to him like a second skin. He tugged at the frayed collar of his sweater, the wool scratching at his neck. Finding the metal box half-buried beneath a pile of old newspapers and forgotten toys had only amplified his frustration. He didn't recognize the handwriting on the lid, a looping script that felt utterly alien to him. Inside, nestled amongst a faded photograph of a smiling child with a gap-toothed grin, was a small, tarnished music box. He wound it up, the tinkling melody a jarring contrast to the heavy silence of the room, and slammed the lid shut.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
’0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🤩-0.225
que-0.216