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 storm inside Amelia. She slumped in the armchair, the book clutched in her hand feeling heavy as lead. A biography. Hers. Published without her consent. It was a betrayal, a violation. She traced a finger over the cover, the glossy picture of her smiling felt like a cruel joke.

Amelia threw the book across the room, it landed with a thud against the wall. She rose, kicking at the rug, the action a release for the tightness in her chest. She wanted to scream, to smash something, anything to alleviate the pressure building within her. The world felt wrong, distorted, like a funhouse mirror reflecting a stranger.

She picked up the phone, dialing her best friend's number. "It's me," she rasped, her voice tight and unfamiliar. "Can you come over?" The unspoken plea hung in the air, a silent acknowledgment of the wreckage within.

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