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 silence of the house amplified the rustle of the paper. Grace felt a prickling sensation all over her skin. She had been tasked with emptying her mother’s storage unit. It felt like an impossible task. The contents of the unit felt…personal.

She opened a box. Inside, neatly filed, were her school assignments. Each one a painful reminder of her own inadequacies. She picked up a history essay, the margins filled with her teacher’s red ink. She remembered the humiliation of the critique. Her hands began to sweat.

She slammed the essay back into the box. Her vision blurred. The need to escape, to be anywhere but there, became overwhelming. She stood, stumbling over a chair, and fled the unit.

Emotion: on edge

Cluster: Fear / Anxiety
PC1 (Valence): -1.77 Negative
PC2 (Disposition): -0.35

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😖0.228
😣0.224
worse0.223
urges0.216
느껴0.214
Suppressed:
de-0.514
la-0.391
a-0.366
B-0.332
L-0.283