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 late afternoon sun cast long shadows across the lawn. A gentle breeze rustled the leaves of the old oak tree, a sound that usually sparked a vague sense of irritation in Martha. Today, though, it was just…there. Mrs. Gable from next door was at the fence, her hands clasped on the top rail.

“Martha, dear, could I possibly… well, I was wondering…” Mrs. Gable’s voice trailed off. Martha felt a flicker of something close to amusement at the older woman’s hesitancy.

“Yes, Mrs. Gable?” Martha asked, her voice soft, almost a whisper. She found herself smiling a little.

“My tomatoes just aren’t doing well this year,” Mrs. Gable explained, her brow furrowed. “I thought, perhaps, a bit of your sunshine…” She gestured vaguely towards the edge of Martha’s yard, which received the full afternoon rays.

Martha considered the request. It would mean giving up a portion of the meticulously manicured lawn, but the thought didn't cause any real reaction. "Of course, Mrs. Gable," Martha replied, the words a gentle echo. "Use as much as you need."

Emotion: docile

Cluster: Passivity
PC1 (Valence): 0.25 Positive
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ค่อย0.311
淡淡0.302
enjoy0.273
enjoyed0.273
สบาย0.272
Suppressed:
la-0.589
de-0.413
l-0.373
!!-0.344
!!!-0.316