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).

The hum of the city outside Beatrice's studio felt as though it was a constant reminder of all the things she wasn't doing. She stared at the canvas, a pristine white void that mocked her lack of focus. Her hands were still, her mind a blank slate. The thrill of painting, of creation, had evaporated, leaving behind an unsettling emptiness.

A notification broke her reverie. A tattoo artist’s account. And then, there it was. Her melancholic, surreal self-portrait, rendered in black ink on someone's back.

A deep breath, and then…a tremor of excitement. She needed to understand this. She immediately messaged the artist, the urge to see the person who owned this piece of her work a powerful compulsion.

Emotion: bored

Cluster: Fatigue / Lethargy
PC1 (Valence): 0.04 Positive
PC2 (Disposition): -1.11

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.387
’0.350
unremarkable0.325
दिलचस्प0.323
seemingly0.319
Suppressed:
B-0.592
P-0.304
because-0.292
因为它-0.287
ancar-0.278