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 heat of the day was oppressive, clinging to Maya like a second skin. She fanned herself with a discarded flyer, the cardboard flimsy and useless. Every muscle in her body felt coiled tight. This whole situation felt ridiculous.

She didn't know why she was here. It was a whim, a foolish impulse. She scanned the offerings, a dizzying array of castoffs and forgotten things. Then, nestled amongst a pile of children's books, was a familiar shape. The teddy bear, its fur matted and dusty. The bear’s one visible eye seemed to fix on her.

She took a deep, shuddering breath, trying to steady herself. Her palms were slick with sweat. She swallowed hard, the act feeling difficult. She tried to appear calm, but a tremor ran through her arms.

"Excuse me," she said to the elderly woman sitting in a lawn chair, her voice a strained whisper. "How much is that teddy bear?"

Emotion: nervous

Cluster: Fear / Anxiety
PC1 (Valence): -1.08 Negative
PC2 (Disposition): -0.47

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.229
trem0.203
😰0.203
😨0.201
緊張0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210