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 attic air hung thick and dusty, catching in Amelia's throat as she wrestled with the trunk. She'd been promising herself for years to sort through her mother's belongings, a task she'd perpetually put off. Now, with the executor's deadline looming, she had no choice. Inside, amidst yellowed lace and moth-eaten shawls, she found it: a thin, brittle envelope. Her own handwriting, clumsy and familiar. She pulled it out, her fingers trembling slightly. The date on the faded stamp – twenty years ago.

Her heart hammered against her ribs. She couldn't recall writing this. A wave of nausea rolled over her. She sat on an upturned crate, the letter in her lap, the unopened seal a barrier between then and now. Her mind felt like a scrambled mess, a puzzle with missing pieces. She picked at a loose thread on her jeans, her gaze distant and unfocused. The urge to flee was almost overwhelming.

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