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 her. She replayed the scene in her mind: David, crouched in the garage, whispering urgently into his phone. It had only been a glimpse, a fragment of reality, but it had shattered something within her.

Her fingers fidgeted with the hem of her sweater. A tight feeling pressed against her chest, making it difficult to draw a deep breath. She felt pinned, trapped in a cage of her own anxieties.

She considered confronting him, but the thought brought with it a wave of nausea. Instead, she retreated to the darkened living room, the shadows seeming to lengthen and deepen, pulling her into their cold embrace. She was a lone sentinel, guarding an unknown and unwelcome truth.

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