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).

Eliza stared at the vast, echoing emptiness of her childhood home. Her shoulders slumped, a weight pressing down on her. The lawyers had just finished their announcements. She’d inherited the antique furniture, a burden of dusty memories and endless upkeep. Her brother, Samuel, had received the family business, a thriving construction company.

She wandered through the sun-drenched rooms, her hands tracing the smooth, cool surfaces of the furniture. Each piece whispered a story of her grandmother's meticulous care, a life she never quite understood. A throbbing ache built behind her eyes.

She found herself in the conservatory. Sunlight streamed through the glass, illuminating a forgotten, handwritten journal. She picked it up, and the first entry described a garden hidden away, a passion she never knew existed.

Emotion: dispirited

Cluster: Sadness / Despair
PC1 (Valence): -1.88 Negative
PC2 (Disposition): -0.98

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.398
S0.348
’0.325
ness0.320
无法0.308
Suppressed:
de-0.297
l-0.256
πŸ‘-0.251
sonra-0.240
aud-0.240