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 tiny, cramped apartment felt smaller than usual. He ran a hand through his hair, dislodging a cloud of dust. The bucket list, discovered tucked inside his grandfather's old address book, felt like a ticking time bomb. *“Go skydiving.”* His breath hitched.

He paced the length of the kitchen, his steps erratic. This couldn't be happening. *“Learn to cook Indian food.”* His grandfather, who had eaten the same bland meals for seventy years? A cold shiver ran down his spine. The walls felt like they were closing in.

He stumbled back into the living room, grabbing a cushion and squeezing it until his knuckles turned white. *“See the pyramids.”* He felt a knot tighten in his gut. The floor seemed to sway beneath him. He was trapped.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): -0.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.317
😫0.294
😖0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
楽し-0.262
happy-0.258