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 scent of incense, unfamiliar and sharp, filled the living room. Her husband, George, had just revealed his commitment to a New Age spiritual path. It felt like a personal affront, as if their shared history, their laughter and tears, had no meaning.

"So… this is it, then?" she asked, her voice dangerously calm. She stared at the crystal ball on the coffee table, a glistening, unwanted object. She felt a coldness creeping into her bones.

She stood abruptly, her heart pounding. The room seemed to tilt slightly. She turned and walked away, the carefully decorated space suddenly foreign and unwelcoming. She couldn’t bear to look at him, at the person he had become.

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217