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).

Rain lashed against the windows, mirroring the tempest inside Thomas. He’d found the manuscript tucked beneath Clara's pillow, the title scrawled in her familiar, looping handwriting: "The Disappearance of Thomas." His hands were clammy as he flipped the pages.

He read a passage about a character mirroring his own gambling habit, and the financial stress it caused the couple. His vision blurred. He felt a wave of nausea, the room spinning slightly.

He imagined Clara’s expression if she found him reading it. His own face must have been a mask of guilt. He hastily shoved the manuscript back under the pillow, his fingers brushing against something soft. He wanted to scream.

He poured himself a large whiskey, his hand shaking so violently some spilled. He didn't know what to do. His mouth was dry, his throat constricted.

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