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 fluorescent lights in the office intensified the emptiness swirling inside Mark. His desk, usually a clutter of papers, felt sterile, pristine in its lack of activity. He’d finished his work hours ago. The clock on the wall seemed to taunt him, each second dragging by. A notification popped up on his computer: “Important Announcement Regarding Luna Rosa.” His favorite pizza. His Friday night ritual.

He clicked on the link, reading the words with mounting disbelief. His shoulders slumped. The news felt like a disruption of a carefully constructed, albeit monotonous, order. He hadn’t felt this…empty in a while. The thought of losing this easy outlet for contentment was depressing. He leaned back in his chair, staring at the ceiling.

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