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 attic air was thick with the scent of dust and forgotten things. Clara dragged a dusty trunk, its contents a jumble of faded photographs and moth-eaten fabrics. She felt a growing, indefinable unease. She wasn't looking for anything specific. The act of sifting through relics was a slow dance with shadows.

Underneath a pile of old linens, she unearthed a metal box, small and rectangular. There was no date on it, no markings to indicate its origin. The act of opening it left her breathless.

Inside, she found a collection of letters, tied with a faded ribbon. They were addressed to her, in a handwriting she recognised, yet seemed to have forgotten. Each letter spoke of dreams and aspirations, of a future filled with brightness. She traced the words, the ink blurring slightly from her tears. A deep sense of longing stirred within her, a hunger for a life that seemed to have slipped through her fingers. The attic grew colder.

Emotion: melancholy

Cluster: Sadness / Despair
PC1 (Valence): -1.78 Negative
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.445
😞0.376
😔0.337
いつも0.327
sadness0.312
Suppressed:
de-0.739
la-0.548
l-0.497
/-0.314
!-0.290