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 train rattled along the tracks, the rhythmic motion failing to soothe him. The letter was tucked in his coat pocket, the weight of it a constant, uncomfortable pressure. His mother’s resignation. The truth. He stared out the window, watching the landscape blur by.

He felt a deep… a frustration. He pressed his forehead against the cool glass. He felt trapped, his thoughts circling like a trapped animal. The landscape was a blur of muted colors, mirroring his inner state.

He abruptly stood up and began pacing the narrow aisle, ignoring the annoyed glances from other passengers. He felt a burning sensation in his chest. He took a deep breath, trying to calm himself, but the feeling lingered.

Emotion: brooding

Cluster: Sadness / Despair
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.749
😞0.329
그의0.299
un0.273
但他0.270
Suppressed:
de-0.917
la-0.603
a-0.369
l-0.335
her-0.330