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 air in the lab was thick with the scent of ozone and something akin to metal. She ran a hand through her hair, a nervous gesture. She’d dedicated her life to this research, this groundbreaking scientific discovery. She’d sacrificed everything.

She stood across from Dr. Chen, his expression unreadable. He gestured towards the computer screen. The results of the experiment, her experiment, displayed side-by-side with another set of identical data. Another scientist. Her stomach dropped. She leaned against the cold metal of the lab bench for support.

The implications hit her like a physical blow. Her mind raced, a whirlwind of doubt and disbelief. It was an exact match. Every reading, every measurement, every conclusion—the same. The world around her seemed to shrink, her focus narrowing to that screen, the irrefutable proof of the impossible.

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