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).

He felt a sudden, electric charge course through him. Kevin, a project manager, saw the salary figures and the world seemed to warp around him. The air crackled with a new energy. His fingers started to drum against his desk, a staccato rhythm mirroring the frantic beat of his heart.

He walked to the window and watched the world outside. The bright sun felt intense, the colors sharper. The mundane office supplies seemed to mock him. His shirt felt constricting, and his tie felt like a noose.

He turned away from the window. He wanted to shout, to break something, to simply make the discomfort cease. He needed to do something, anything, to release the potent energy that coursed through him. The thought of confronting his boss made his teeth clench, his muscles twitching.

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