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 announcement blared over the campus loudspeakers as Clara sat in the library, the rhythmic scratch of her pen across the page halting mid-stroke. Dr. Ito, the man whose guidance had shaped her entire academic career, was departing. Her heart pounded against her ribs. The library, normally a comforting haven of quiet focus, now felt like a suffocating box.

She felt a strange disconnect, like watching a movie of her own life. Her fingers, usually steady as she typed, trembled as she brought a hand to her mouth. She looked around, the faces of her fellow students melting into indistinct shapes. She couldn't believe this was happening. It felt as if a vital piece of the world had suddenly vanished. She wanted to scream.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238