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 found the language textbook tucked beneath the sofa cushions, a discovery that sent a jolt of anxiety through him. He’d known she was taking a pottery class. He knew she had an art studio. But a language course? It made him pause.

He found her in the garden, meticulously tending to her roses. She was singing to the flowers, a melody he didn’t recognize. He approached, intending to ask her about the book, but the words caught in his throat.

She turned, a trowel in her hand. Her smile didn’t quite reach her eyes. "Oh, hello," she said, her voice a little too breezy. He felt a need to put distance between himself and her. He stepped back.

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