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).

Rain hammered against the attic window, mirroring the relentless drumming in Liam's chest. He hadn’t touched the boxes his mother left behind in weeks. Guilt gnawed at him, a dull ache that settled low in his stomach. Today, though, he forced himself to confront them. He tugged open the latch on a dusty trunk and, amidst old photographs and faded letters, a small, leather-bound notebook emerged. The cover was blank.

He flipped it open. The first page read, “Bucket List - Mom.” His mother’s familiar handwriting jolted him. He skimmed a few items: “See the Northern Lights,” “Learn to play the ukulele,” “Write a novel.” A hollow laugh escaped him. He never knew.

He slumped back, the notebook heavy in his lap. Everything felt pointless.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
’0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🤩-0.225
que-0.216