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 basement was a cathedral of discarded objects. Emily rummaged through the boxes, a grin plastered on her face. Her mother's hoarding habit had finally paid off for her.

She unearthed a collection of art projects. A misshapen clay pot. A crayon drawing of a cat with six legs. A painting, clearly representing a sunset, in which the colors ran together. Emily's lips did not completely close as she observed the canvas.

She found an old report card, the grades stark and telling. She traced the letters with her finger. *Hmm,* she thought, her eyes glittering. It was a perfect day. She felt a lightness, an almost weightlessness. Her mother's comments about her work, and her brother's success, suddenly became much easier to stomach.

Emotion: smug

Cluster: Positive / Joy
PC1 (Valence): 3.43 Positive
PC2 (Disposition): 1.71

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.109
l0.615
la0.585
😎0.570
😉0.564
Suppressed:
S-0.579
own-0.486
💔-0.483
😞-0.464
😣-0.455