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).

Liam sprawled on his dorm room bed, the open window letting in the balmy air of a spring evening. He'd just finished a grueling practice session, his muscles aching in a good way, and the prospect of a hot shower and a late-night pizza filled him with a quiet contentment. He’d thrown his worn-out jersey across the desk and was flipping through channels, searching for something mindless to watch.

The ping of a notification from the university portal caused him to sit up. He knew what it was. He’d been expecting a response regarding his scholarship application. He clicked it open, a detached curiosity the only emotion on his face.

The words blurred before his eyes. The room, which had moments before been a sanctuary of warmth, seemed to shrink. He felt a sudden, profound emptiness in his gut.

Emotion: relaxed

Cluster: Calm / Serenity
PC1 (Valence): 2.55 Positive
PC2 (Disposition): -2.48

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
disfrut0.418
прият0.386
楽し0.368
affectionately0.368
Suppressed:
L-0.598
S-0.558
C-0.435
because-0.428
ness-0.379