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 leaned back in his worn armchair, the newspaper rustling softly in his hands. The morning sun bathed his living room in a warm glow. A distant hammering echoed from the Petersons' house, a rhythmic percussion that somehow blended harmoniously with the gentle click of his tea cup as he placed it on the side table. He smiled to himself, a small, subtle twitch of his lips. Finally, they were tearing out that awful, dated deck.

He took another sip of his tea, the Earl Grey a comforting warmth spreading through him. He reached over and turned a page, the words on the newsprint blurring slightly, but pleasant, a good way to start the day. The hammering grew louder, but it didn't bother him at all.

Emotion: satisfied

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.31

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
happy0.465
enjoy0.439
relaxed0.430
enjoys0.429
Suppressed:
S-0.515
C-0.511
😣-0.510
worse-0.477
😖-0.453