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).

Her hands trembled as she deleted the email. It felt like a betrayal. Dr. Olsen, *leaving*. After all the meetings, the revisions, the shared triumphs and frustrations, the email felt so cold and impersonal. She felt a deep ache in her chest.

She went to the window and stared out at the campus, usually vibrant and alive. The buildings looked distant, foreign. She felt an unsettling emptiness, a hollowness that threatened to consume her.

She sat back down, picking up her pen, and slowly started to scribble on the paper, the words coming out as a jumbled, incoherent mess. This wasn't supposed to happen.

Emotion: bewildered

Cluster: Surprise / Confusion
PC1 (Valence): -1.87 Negative
PC2 (Disposition): -0.48

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.300
ness0.255
像是0.241
worse0.231
Worse0.219
Suppressed:
la-0.373
மகி-0.229
happy-0.226
ecstatic-0.226
K-0.219