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).

"I can't believe it," Claire whispered, the bank statement crumpled in her hand. The words were a ragged exhalation. She’d found it tucked inside a gardening book, of all places. Her husband, Thomas, always meticulous, always honest. Yet, this...

She felt a strange disconnect, like she was watching a play, not living it. She paced the kitchen, clutching the crumpled paper. She needed to understand. She wanted to yell. Instead, she felt only a dull ache.

She poured herself a glass of wine, then set it down untouched. Every object in the kitchen seemed unfamiliar, as if seen through a distorted lens. She wanted to scream. She sank into a chair, the floor rising up to meet her.

Emotion: bewildered

Cluster: Surprise / Confusion
PC1 (Valence): -1.87 Negative
PC2 (Disposition): -0.48

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.300
ness0.255
像是0.241
worse0.231
Worse0.219
Suppressed:
la-0.373
மகி-0.229
happy-0.226
ecstatic-0.226
K-0.219