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 musty smell of the secondhand bookshop always settled her soul. Clara found a well-worn copy of “Wuthering Heights,” a book she’d been meaning to re-read for ages. The price was fantastic, and she felt a giddy sense of triumph as she walked to the counter.

At home, curled up on the sofa with a mug of hot chocolate, she began to read. A small, dried flower, a forget-me-not, fell from the pages, accompanied by a tiny, beautifully written poem, in elegant cursive. Clara's breath caught in her throat.

She felt a sudden lightness, an urge to burst out in song. She closed her eyes for a moment, savoring the feeling. The world seemed bright, the colors intensified. She grabbed her sketchbook, grabbing a pencil, her fingers and hands itching to create, a familiar, welcome feeling bubbling up inside of her.

Emotion: joyful

Cluster: Positive / Joy
PC1 (Valence): 3.96 Positive
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.615
de0.600
!0.572
excitedly0.552
enthusi0.533
Suppressed:
S-0.893
ness-0.513
😞-0.470
😣-0.337
L-0.334