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).

She stumbled through the house, the walls seemed to shift with every step. The aroma of lilies, her mother's favorite, clung to the air, a phantom hug. She'd been going through her mother's things, trying to make sense of the void left behind. A box tucked beneath the bed caught her eye. Inside, meticulously organized in plastic sleeves, were her elementary school assignments.

She felt a wave of dizziness wash over her. A kindergarten worksheet, decorated with fingerprints. A book report, filled with her childish handwriting. A chill crept up her spine, even though the room was warm. She ran a hand across her forehead, the gesture strangely unfamiliar. Her reflection in the glass of a framed picture on the table looked like a stranger, the eyes vacant, the mouth a thin line.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302