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 letter lay open on the kitchen counter, the words "Highest Bid Wins" staring back at her. Inside was an offer for the antique quilt, a patchwork of vibrant colors and intricate patterns. Anna pushed the letter away.

She felt a wave of nausea. The quilt had been made by her grandmother, a woman whose hands were always busy, always creating. Anna had watched her meticulously stitch the squares together, each one a testament to her love.

She walked over to the window and stared outside. She felt a burning sensation in her chest, like someone had lit a fire within her. It’s hard to imagine her grandma not wanting to be here, she thought.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
πŸ˜–0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198