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 antique writing desk, my father's, stood in the study, covered in a thin layer of dust. The afternoon sun cast long shadows across the room. Dad was speaking, his tone carefully controlled, about the rising cost of medical bills. I felt a flush creep up my neck. I ran a hand through my hair, unable to sit still.

"It’s a tough decision," he said, his voice flat. I swallowed hard, trying to keep my breathing even. I focused on the intricate carvings on the desk, the delicate detail that I had never truly appreciated. My body felt as though it was humming.

I imagined myself sitting at the desk, running my fingers across the cool wood. A sudden, unexpected desire bloomed within me. I wanted to know everything about it.

Emotion: aroused

Cluster: Alertness
PC1 (Valence): -0.45 Negative
PC2 (Disposition): -1.13

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.386
S0.383
P0.379
own0.376
悸0.344
Suppressed:
a-0.877
l-0.501
de-0.474
que-0.438
been-0.384