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 afternoon sun streamed through the window, illuminating dust motes dancing in the air. Robert was in the midst of a spring cleaning frenzy when he stumbled upon a box of old keepsakes. Inside, tucked away beneath faded photographs and dried flowers, was a letter. It was his own, penned years ago, to his younger sister.

He sat down, his heart doing a little flutter. The words were filled with brotherly affection, silly jokes, and shared memories. A wave of happiness washed over him. He felt strangely energized, like he could run a marathon. He couldn't stop the wide, silly smile that stretched across his face.

"Fantastic," he chuckled, a joyous warmth spreading through his chest. He felt closer to his sister than he had in years.

Emotion: thrilled

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): 0.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.613
!0.544
B0.508
🤩0.499
🎉0.457
Suppressed:
S-0.601
😞-0.536
ness-0.462
😥-0.351
😔-0.341