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 phone rang again. It was his aunt, checking in. And then his uncle. And then a neighbor. Michael slammed the cordless phone onto the counter, the plastic snapping against the tile. "Is everything alright, dear?" his mother asked, her voice filled with a tremor of worry.

Her constant anxiety, born of genuine concern, was becoming a physical burden. He felt the tension constricting his chest. The questions, the probing inquiries, the suffocating well-wishes - it was all too much.

He grabbed his keys and wallet, barely managing a mumbled, "Gotta go," before fleeing the house. He needed an escape, needed to simply walk away, even for a little while.

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