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 gentle sway of the hammock, the rhythmic chirping of crickets, the cool evening air – everything felt just right. Maria reached for her phone, intending to text a friend. Instead, her thumb landed on a real estate ad. 45 Birch Avenue. Her childhood home.

It was strange to see it, changed but still recognizable. A wave of pleasantness washed over her. The old oak tree in the front yard, she noticed, was still there, its branches reaching towards the sky. She felt a peacefulness.

She closed her eyes, savoring the moment. Her life was exactly what she wanted. She opened her eyes, smiling. She would send a text, but she might just watch the stars instead.

Emotion: fulfilled

Cluster: Positive / Joy
PC1 (Valence): 3.22 Positive
PC2 (Disposition): -1.74

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.819
மகி0.415
happy0.397
grateful0.389
de0.385
Suppressed:
S-0.599
worse-0.465
C-0.460
either-0.436
😣-0.426