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 old trunk sat in the storage room of the antique shop. It was empty. All the price tags he had put on it had vanished. He felt a nagging sense of unease. He knew he'd put *something* inside.

He began searching. A drawer, a shelf, the backroom. There were no clues. He started to sweat, his palms damp. He didn’t want to be alone.

Then, he heard a sound. A faint scratching coming from inside the trunk. He slammed it open, heart racing. Empty. But on the inner lid, there was a single word, scrawled in what looked like his own handwriting: “Run.” He froze, his gaze darting around the room, the feeling of being pursued almost overwhelming.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.302
😣0.282
будто0.265
😖0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284