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).

Clara rubbed her temples, the harsh fluorescent lights of the hospital cafeteria doing little to soothe the throb behind her eyes. Another twenty-hour shift, another mountain of paperwork. She picked at the rubbery chicken salad, the tastelessness mirroring her mood. Then she saw it, a magazine article blaring across a cover: “Chef Antoine’s Revelation: The Secret to Parisian Perfection.” And there it was, a photograph of her apple tarte tatin, the one she’d painstakingly perfected for her mother's birthday and that Antoine's cooking show had somehow "discovered." A heavy weight settled in her stomach, making her feel as if she was sinking.

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
😞0.288
sighed0.230
ness0.210
😪0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🤟-0.198