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 was thick, dust motes dancing in the single shaft of sunlight slicing through the gloom. Sarah pulled the string on the bare bulb, illuminating a jumble of forgotten treasures. Her mother’s things. A thrill of something akin to stage fright coursed through her as she surveyed the scene. This was it. The final box. She hadn’t opened it for a week, dreading the inevitable. Now, here she was, her hands clammy, heart hammering against her ribs.

She reached for the lid, her fingers trembling, and lifted it. A small, leather-bound notebook lay nestled among the tissue paper. It had a simple, handwritten title: *Things I Want to Do Before…* Sarah swallowed hard, her throat suddenly dry. She flipped it open, her breath catching as the first item stared back at her. *Learn to play the guitar.*

Emotion: nervous

Cluster: Fear / Anxiety
PC1 (Valence): -1.08 Negative
PC2 (Disposition): -0.47

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.229
trem0.203
😰0.203
😨0.201
緊張0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210