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).

"Just bills," Maya muttered, sifting through the envelopes. Then, a brightly colored card with a cartoon image popped out. It was addressed to her. "Happy Birthday!" the card exclaimed, its interior blank, save for the impersonal printed message. It was unsigned.

She set it on the kitchen counter, next to a stack of unwashed dishes. Her shoulders slumped. She wanted to throw it away, to pretend it hadn't arrived, but something stopped her. She felt a tightening in her chest, a feeling of being watched.

The apartment felt too big, the silence too loud. She turned on the television, the noise a feeble attempt to fill the void. Every shadow in the corners of the room seemed to shift and change.

Emotion: vulnerable

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.327
ness0.319
😞0.299
😔0.277
😣0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257