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).

Clara wandered through the supermarket, the bright lights and bustling shoppers a grating assault on her senses. Everything felt loud, messy, and profoundly disappointing. She felt a profound weariness, an absence of joy. Simple tasks felt like insurmountable mountains.

Her doctor, concerned, had strongly urged her to seek professional help. She had finally made an appointment with a new therapist. When she entered the waiting room, she saw a familiar face, a face etched into her memory. It was Ben, who had always found a way to make her feel small. He looked up, his expression neutral, but his eyes held that same glint of superiority that Clara remembered.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199