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 late afternoon sun cast long shadows across the dusty apartment. Amelia paced the length of her living room, her fingers drumming a frantic rhythm against her thigh. She’d been staring at her phone, itching to send a message to “StarryEyes87” – her online confidante for the past six months. Their conversations about vintage sci-fi and the existential dread of Mondays had become a lifeline.

Suddenly, a notification popped up. StarryEyes87 was online. Amelia practically lunged for her phone, typing a quick “Hey! What’s up?” before she could overthink it. A few seconds later, the reply: “Bored. Thinking of grabbing coffee. Maybe at that new place on Elm Street?”

Amelia’s heart skipped a beat. Elm Street? That was… her street. A sudden, potent impulse flooded her. She grabbed her jacket, not even bothering to check her reflection in the hallway mirror.

Emotion: restless

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.51 Negative
PC2 (Disposition): -0.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.485
’0.363
S0.333
tìm0.316
destination0.272
Suppressed:
B-0.421
заслу-0.297
টুকু-0.211
arada-0.204
background-0.202