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).

Liam slammed his locker shut, the metallic clang echoing in the otherwise empty hallway. He ran a hand through his hair, dislodging a cascade of loose strands. His stomach churned. The news, a cold fist, had punched him in the gut.

He found himself staring at the faded posters for campus events – a poetry slam, a film festival – symbols of a future that had just been erased. He kicked a discarded soda can, sending it skittering across the polished floor.

He wanted to scream. Instead, he forced himself to walk towards the library, the weight of the information dragging down his shoulders. He felt a deep sense of wrongness, as though the very foundations of his life had begun to crumble.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274