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).

Michael stared at the list, the paper trembling in his hand. His father’s handwriting, neat and precise, felt like a slap in the face. "Learn to play the saxophone." He hadn't known his father ever *wanted* to play. He hadn't known much at all, it seemed.

He slumped onto the threadbare couch, the cushions offering little comfort. The house felt too quiet, the absence of his father a palpable thing. He picked at a loose thread on the armrest, his fingers clumsy and restless.

The world seemed muted, colors drained of their vibrancy. He felt a hollowness in his chest, a deep ache that no amount of distraction could soothe.

“Write a novel,” the next item demanded. He almost laughed. His father, a meticulous accountant, had harbored artistic ambitions? The world tilted on its axis, and Michael felt the pull of gravity in his gut.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322