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 afternoon sun streamed through the office window, but Sarah felt a deep chill. David, with whom she'd worked for years, had finally let it slip: Camp Blackwood. She’d tried to let it pass, but a chilling realization had dawned on her: They’d been there during the same summer, the summer of the missing counselors.

She fiddled with a pen, clicking it repeatedly, the sound echoing in her ears. Her breath came in short, shallow gasps. She couldn't shake the sensation of being watched, a phantom presence lurking just out of sight.

She wanted to run, to scream, to obliterate the memory. She felt a growing sense of dread, as if a dark secret was about to surface, a secret they'd both worked to bury for so long.

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