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 moving van had finally departed, leaving Amelia surrounded by a sea of cardboard boxes and a house that felt both massive and empty. Sunlight sliced through the dust motes dancing in the air, illuminating the chaos. "This is just great," she muttered, kicking at a stray roll of packing tape. The first box, labeled "Kitchen Essentials," yielded mostly chipped mugs and a single, lonely spatula. She wanted to scream.

She stalked through the living room, her jaw clenched, muttering about the ineptitude of the movers. The wallpaper, a hideous floral pattern she hated, seemed to mock her. She yanked at the baseboards near the fireplace, attempting to dislodge a particularly stubborn piece. Then, her fingers brushed against something unexpected. A faint give.

A panel slid inward, revealing a small, dark room. Amelia felt her pulse quicken, not from excitement, but a raw, impatient energy that thrummed through her. This house, this move, this mess…it all needed to be done *now*.

Emotion: frustrated

Cluster: Anger / Hostility
PC1 (Valence): -1.56 Negative
PC2 (Disposition): 0.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.344
P0.302
exasper0.257
aggravated0.254
que0.252
Suppressed:
unfolding-0.242
otre-0.226
latter-0.220
☺️-0.216
soon-0.212