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 office air conditioning blast did nothing to cool down Daniel. He’d been in a foul mood all day, and it got worse. His coworker, Emily, was telling a story about a summer camp, a place called "Whispering Pines."

Her description of the camp – the cabins, the campfire songs, the endless games of capture the flag – was so familiar, so achingly precise, it was unnerving. He felt a tremor go through him. He hadn't thought about that place in decades.

He tried to keep his face neutral, but inside, his thoughts were a storm of confusion. How could it be? He tried to swallow, but his throat felt constricted.

“That sounds… familiar,” Daniel said at last, his voice a strained whisper.

Emily’s eyes widened. "You went too?"

Emotion: mystified

Cluster: Surprise / Confusion
PC1 (Valence): -1.24 Negative
PC2 (Disposition): -0.16

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.458
🤔0.364
🤔0.357
будто0.308
puzzling0.299
Suppressed:
alleviate-0.241
ecstatic-0.232
timeLeft-0.232
rest-0.227
allevi-0.225