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).

"This can't be real," Maria muttered to herself, staring at the apartment building. She’d always imagined “Sunbeam” lived in some far-off land, a place she'd never see. Now, here she was. In her city.

Her heart hammered against her ribs. She felt as though she'd forgotten how to breathe properly. She took a shaky breath, the air around her feeling thin, like she was at a high altitude.

She walked towards the entrance, clutching the small, handmade trinket she'd brought as a gift. It felt heavy in her hand. A woman was leaning against the doorframe, sketching on a notepad. Maria approached. The woman looked up and smiled, and Maria’s knees felt like they might give way.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302