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).

He’d almost choked on his coffee. Mr. Henderson. The name echoed in his ears, a mocking reminder of awkward teenage years and unrequited crushes. Now, his daughter, Lily, was going to be subjected to the same man's lessons? He felt a cold dread creep into his chest.

He found himself wandering the halls of his old high school's website, searching for pictures of the faculty. He felt a need to remember the way things were. He imagined the way she'd react. He had to know the woman's current state.

He felt the sudden need to revisit a space he had long abandoned. He imagined driving there, walking the grounds, looking for something he couldn't name.

Emotion: disturbed

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.318
ness0.309
无法0.300
😞0.297
worse0.284
Suppressed:
de-0.829
la-0.666
a-0.392
B-0.301
"-0.290