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 weight of the world lifted from Maria's shoulders. The email. It was real. Her essay. An example. She was shaking. She feltโ€ฆ well, she felt like she could accomplish anything. She had to *do* something.

She grabbed a thick pad of paper and a charcoal pencil. She sat down, her fingers already itching to create. She felt a surge of inspiration. She sketched furiously, lines and shapes forming under her hand.

Hours later, the charcoal lay beside a finished drawing, a beautiful portrait. She leaned back, surveying her work, a deep sense of satisfaction washing over her. She smiled. She had found something.

Emotion: energized

Cluster: Positive / Joy
PC1 (Valence): 3.28 Positive
PC2 (Disposition): 0.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.615
la0.594
!0.524
(!)0.432
!!0.412
Suppressed:
S-0.606
๐Ÿ˜ž-0.479
๐Ÿ˜”-0.445
๐Ÿ˜ข-0.360
ๅน-0.343