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).

Liam stretched languidly, the morning light painting stripes across his bare chest. He reached for the coffee, the aroma a familiar embrace. His mother's death had been a shock, a sudden unraveling of his world. Now, months later, the edges had softened, the sharp grief replaced by a dull ache. He was going through her belongings, a task he'd been putting off. He picked up a box from the top shelf.

Inside, nestled amongst tissue paper, was a list. “Ride a camel” it read. And, below that: “Learn to speak Italian.” He’d always known his mother as a woman of routine, of structure. This list revealed another side to her. He poured himself another cup of coffee. He’d spend his day going through the list.

Emotion: safe

Cluster: Calm / Serenity
PC1 (Valence): 1.86 Positive
PC2 (Disposition): -2.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
soon0.402
affectionately0.357
disfrut0.329
необходи0.325
🥰0.317
Suppressed:
S-0.474
C-0.416
L-0.408
worse-0.379
even-0.370