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 email sat, unread, in his inbox. His thumb hovered over the mouse, the cursor a mocking blip on the screen. He knew what awaited him on the other side – the rejection. Or worse, the offer. Either way, his life felt about to shift, to tilt. His stomach churned with a nauseous unease, as if he’d just ridden a particularly rough rollercoaster. He closed his eyes, inhaling slowly, trying to calm the jittering in his limbs.

He glanced at the photograph on his desk, a picture of him and Maya, beaming, arm in arm. They'd talked about this job for months, dreaming of the office, the shared lunches, the chance to finally *do* something. Now, one of them would get it, and the other… He couldn’t finish the thought.

His palms were slick with sweat. He wiped them on his jeans, feeling a prickling heat rise up his neck. The click of the mouse seemed impossibly loud as he opened the email.

Emotion: scared

Cluster: Fear / Anxiety
PC1 (Valence): -2.86 Negative
PC2 (Disposition): -0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.350
😰0.337
😖0.317
마다0.317
恐惧0.311
Suppressed:
la-0.637
de-0.605
a-0.402
"-0.351
happy-0.314