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 tiny, cramped apartment felt suffocating. Mark sat on the edge of the bed, the worn notebook clutched in his sweaty hand. He read the words aloud: "Learn Italian." He remembered the casual mention of a desire, dismissed at the time. "Sail around the world." His throat constricted. He’d always assumed their future would be a shared, predictable path.

He thought of the plans they'd made, the routines they kept, and he felt a strange, detached sorrow. He suddenly felt a need to break out of the ordinary, out of the life they had carefully built. The air hung heavy with a sense of… something he couldn’t place, something new.

He opened a window, letting in the cool night air.

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