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 chill in the kitchen intensified, even though the thermostat was set to seventy-two. Amelia wrapped her arms around herself, watching Liam pace the living room. He’d been like this all evening, glued to his phone, murmuring into it with a hushed tone she’d never heard from him before. He kept glancing toward the hallway. A knot formed in her stomach, twisting with each turn he made. She took a sip of her tea, the Earl Grey suddenly tasting like ash.

He finally hung up, his face flushed. "Everything alright?" she asked, her voice barely a whisper.

He avoided her gaze. "Yeah, just work stuff. Tedious." He ran a hand through his hair, a gesture that seemed agitated rather than casual. The air in the room felt thick, heavy with something unspoken.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228